TW201009328A - Image data processing apparatus and method for defect inspection, defect inspecting apparatus and method using the image data processing apparatus and method, board-like body manufacturing method using the defect inspecting apparatus and method - Google Patents

Abstract

Defect candidates are extracted from among photographed images of a board-like body by using a signal threshold value. From among the extracted defect candidates, the defect candidates at the same position in the width direction are searched, and intervals between the position of the searched defect candidate in a transfer direction and the position of the adjacent defect candidate in the transfer direction are repeatedly obtained, thus, generation frequency distribution of the intervals is obtained. When the interval generation frequency of interest is over a frequency threshold value, it is discriminated that the board-like body has periodic defects in the transfer direction. The frequency threshold value is so set as to substantially increase as the interval of interest becomes shorter.

Description

[Technical Field] The present invention relates to a processing apparatus and a processing method for image data for defect inspection for inspecting defects existing in a transparent plate-like body such as a glass plate, and respectively A defect-inspecting apparatus, a defect-inspecting method, a method of manufacturing a sheet-like body using the same, and a recording medium readable by a computer for recording a program for processing an image data for performing defect inspection. [Prior Art] Currently, since glass sheets are used in electronic devices such as flat panel displays and thin film solar cells, there is a strong demand for glass sheets having thinner thicknesses, bubbles or scratches, and few or no defects. As a defect existing in the glass plate, a flaw formed on the surface of the glass plate can be cited. For example, in the floating method, a long plate-like body having a fixed thickness is taken out from a melting furnace and conveyed on a driving roller. At this time, the surface of the glass plate is damaged by foreign matter adhering to the driving roller or minute projections on the driving roller, thereby causing a flaw. Since a plurality of driving rolls are used for the conveyance of the glass sheets, there are many opportunities for occurrence of minute scratches on the surface of the glass sheets. Since the above-mentioned flaws are periodically generated, since the prior art, not only the glass sheets but also the method of inspecting the intermediate defects in the strip shape, various methods for inspecting the periodic defects have been proposed in the inspection in the step. Patent Document 1 discloses a measurement method in which the period of the defect existing in the object to be inspected is measured, and the defect data and the normal data are binarized, and the distance between the defect data is obtained. Each distance is calculated into the 141812.doc 201009328 line frequency calculation to calculate the periodic component of the distance, and the basic period of the defect is extracted from the periodic component. Patent Document 2 discloses a method of classifying a defect detected by photographing an image data sheet obtained by an object to be examined as a periodic defect and a non-periodic defect' and inspecting a periodic defect. [PRIOR ART DOCUMENT] [Patent Document 1] Japanese Patent Laid-Open Publication No. Hei 7-86474 [Patent Document 2] Japanese Patent Laid-Open Publication No. Hei. No. Hei. However, in Patent Document 1, when the periodic component of the distance between the defect data is calculated, since the defect data and the noise data cannot be distinguished, it is difficult to efficiently calculate the cycle amount. When the defect is small, it is necessary to lower the threshold of binarization. Therefore, the amount of processing of the noise data as the defect data becomes extremely large, and the accuracy of calculation of the periodic component is further lowered. In the other aspect, in Patent Document 2, when the defect detected in the image data obtained by photographing the object to be inspected is classified as a non-periodic defect of the job title, the image data is binarized, and The size of the area obtained by binarization which is regarded as the part of the defect is higher than the specific value of the classification period defect. Therefore, the sputum product k ～ will be misidentified as a periodic defect by the part of the defect that is extremely small and is considered to be caused by the noise. That is, 'the defect caused by the driving roller or the like on the glass plate is small', so when the image data is binarized, it is difficult to reduce the threshold value in order to reduce the defect of 2 141812.doc 201009328. The noise data is distinguished from HI: 'It is extremely difficult to discriminate the periodicity of defects caused by small scratches caused by driving rollers and the like.

In order to solve the above problems, an object of the present invention is to provide a defect that can detect the presence of a periodic defect even if a captured image contains a noise component when detecting a defect in a plate-like body such as a glass plate. Processing device and processing method for inspection image data, defect inspection device and defect inspection method using the same, use of the inspection method or the inspection method of the defect inspection device, and recording of image data for performing defect inspection A recording medium readable by a computer for the processing method. [Means for Solving the Problems] In order to achieve the above object, the present invention provides a processing apparatus which is characterized in that it uses a surface to relatively move a plate-like body in a specific direction to face a plate-like body. a processing device for inspecting image data for defect inspection of a defect existing in the plate-like body; and a processing unit for extracting a plurality of defect candidates from the image using the first signal threshold And searching for the defect candidates having the same position in the width direction orthogonal to the specific direction, that is, the moving direction, from the plurality of extracted candidate candidates in the moving direction, and determining the candidate detected by the search: The interval between the position in the moving direction of the plate-like body and the position of the defect candidate in the moving direction adjacent to the detected defect candidate in the moving direction is obtained by repeating the above processing to obtain a plurality of intervals And determining a frequency at which the plurality of intervals are generated, and determining that the board is when the frequency of occurrence of the interval of interest exceeds the set frequency threshold Shape 1418I2.doc 201009328: The moving direction i has a periodic defect; the frequency threshold used in the processing unit is defined according to the interval of the above-mentioned attention, and when the two frequency thresholds are defined as different values, The above-mentioned frequency threshold is set in such a manner that the above-mentioned attention interval of the frequency 闺 value of the larger one is smaller than the above-mentioned attention interval of the frequency threshold of the smaller one. According to a second aspect of the invention, there is provided the processing device of the first aspect, wherein the processing (β) includes a reference table that indicates a relationship between a density of the candidate candidate and the first signal: a value of the defect candidate. In the manner in which the density of the two eyes is privately generated, the above-mentioned reference table is used to set the value of the first value of the above-mentioned number. The frequency threshold value is a value that varies according to the value of the target generation density in addition to the change according to the above-mentioned attention interval. . The third aspect of the present invention provides the processing apparatus of the above-described type 1 or type 2, wherein the frequency threshold used in the processing unit is regulated in such a manner that the false noise component is randomly distributed in the area. And the noise component having the position in the width direction at the same position as the defect candidate 'analytically obtains the frequency of generation of the noise component with respect to the interval' or a simulated image formed by the noise component The image of the noise © component is used as a defect candidate, and the frequency of occurrence of the noise component with respect to the interval is determined. The frequency threshold is defined based on the obtained generation frequency. Mode 4 of the present invention provides the processing apparatus of the above-described Type 3, wherein the noise component is generated such that the density of the noise component in the image varies depending on the area of the image. According to a fifth aspect of the present invention, there is provided a processing apparatus according to any one of the above aspects 1 to 4, wherein the processing unit searches for a search target of the defect candidate in the width direction and the moving direction. Forming a plurality of unit regions having the same size by dividing into a plurality of portions, and when a plurality of unit regions including a plurality of defect candidates are at the same position in the width direction, the defect candidates are set to be in the width direction The positions are the same, and the above interval and the above-mentioned generation frequency are obtained. According to a sixth aspect of the invention, there is provided the processing device of any one of the above-mentioned aspects 1 to 5, wherein a frequency distribution in a width direction indicating a distribution of the generation frequency in the width direction is obtained, and the width direction is used. The unevenness of the above-described generation frequency in the above-described width direction in the frequency distribution is generated, and the defect generation pattern is classified. According to a seventh aspect of the present invention, the processing apparatus according to any one of the above aspects 1 to 6 wherein the said plate-shaped system has a continuous shape in the moving direction, the processing unit divides the plate-shaped body into The region of the plate-shaped body having the set length is subjected to the above-described discrimination for a plurality of units as the inspection target of the unit. According to a seventh aspect of the invention, the processing device of the seventh aspect, wherein the processing unit records a generation frequency distribution of the interval defined by the interval and the position in the width direction for the plurality of time series units, according to the recorded The generation frequency distribution defines the frequency of occurrence and the position in the width direction to determine the generation frequency, and expresses the generation frequency as time-series data, thereby displaying the defect generation information on the screen. The mode 9 of the present invention provides the processing apparatus of the above-described type 8, wherein the plurality of times of the time series data of the generated frequency are changed by at least one of the positions of the above-mentioned attention intervals 141812.doc 201009328 and the direction of the direction. When the frequency is generated, the 'preface' data is overwritten in the same graph and the force σ is displayed in the picture by 7]λ. The processing apparatus according to any one of the above aspects 1 to 9, wherein the processing unit searches for a defect candidate having the same position in the width direction in the moving direction, and The neighboring candidate candidate obtains the interval in the moving direction as the previous defect candidate, and also obtains the interval between the plurality of preceding defect candidates in the moving direction, and obtains the plural by repeating the above processing. For the interval, the frequency of the complex (four) interval is extracted, and when the frequency of occurrence of the interval of interest exceeds the set frequency threshold _, it is determined that the plate-like body has a periodic defect in the moving direction. The present invention provides a processing apparatus according to any one of the above aspects 1 to 10, wherein the processing unit is configured to refer to an interval in which the periodicity is determined in the moving direction as a pitch interval. Further, it is defined that the defect having the pitch interval described above is in the region of interest of the position where the width direction is located, and the second signal threshold is used to extract detailed defect candidates from the image of the region of interest from the beginning of the image. a search area centered at a position at which the detailed defect candidate obtained from the extraction in the moving direction is separated from the pitch interval, and the detailed defect candidate is searched for using the second signal threshold in the search region, and the search is evaluated separately The detailed defect candidate detected and the attribute of the detailed defect candidate obtained by the extraction are determined based on the evaluation result whether or not the region of interest includes the periodic detailed defect candidate in the moving direction. In a processing apparatus according to any one of the above aspects 1 to 11, wherein the processing unit searches for and detects a defect candidate having the same position in the width direction in the moving direction and seeks When the interval is exceeded, the attribute of the detected defect candidate or the similarity between the defect candidate and the specific defect candidate is evaluated. The above interval is obtained when at least one of the attributes and the similarity satisfies the set condition. A form 13 of the present invention provides a defect inspection apparatus characterized by:

A person who inspects a defect existing in a plate-like body includes a light source that emits light toward a surface of the plate-like body, and a camera that moves relative to the plate-like body with respect to the light source on one side. An image of a plate-shaped body that is irradiated with light by the light source; and a processing device according to any one of the above aspects, wherein the processing unit of the processing device photographs from the camera using the first signal threshold Extracting the plurality of defect candidates from the obtained image, and searching for the direction of relative movement of the camera relative to the plate-shaped body from the plurality of extracted defect candidates in the moving direction, that is, the moving direction The candidate candidates having the same position in the width direction are orthogonal to each other. The type 14 of the present invention provides a processing method characterized in that it detects the defects existing in the above-mentioned plate-like body by using the image obtained by moving the plate-like body relative to the surface (4) in a specific direction. a method for processing image data for defect inspection; preparing a first signal threshold value, extracting a plurality of defect candidates from the captured image; searching for the above-mentioned special W candidate from the extracted plurality of defect candidates in the moving direction The 疋 direction, that is, the defect in which the moving direction is orthogonal to the position in the width direction is the same as the candidate in the moving direction, and the position of the defect candidate detected in the moving direction by the search 141812.doc 201009328 is not present in the moving direction. The interval between the positions of the candidate candidates adjacent to the defect candidate in the moving direction is obtained by repeating the above processing to obtain a plurality of intervals; determining the frequency of generation of the plurality of intervals; when the frequency of the interval of attention exceeds When the frequency threshold is set, it is determined that the plate-like body has a periodic defect in the moving direction; the frequency threshold is based on the interval of the above-mentioned attention To be provided, when the two frequency thresholds are different, so that the above-described predetermined frequency interval greater attention as one of the threshold values is less than a predetermined frequency smaller one embodiment the threshold value above attention interval, setting the frequency threshold value. The mode 15 of the present invention provides the processing method according to the above aspect 14, wherein the inspection condition is set before the discrimination is performed; and in the step of setting the inspection condition, the density of the defect candidate is set to the target generation density. In the mode, the second signal threshold is set using a reference table; the frequency threshold is a value that varies according to the value of the target generation density in addition to the change in the interval of interest. The mode 16 of the present invention provides a processing method of the above-described Type 14 or Type 15, wherein the frequency threshold is defined in such a manner that an image of the (four) component of the analog image formed by the noise component is formed. As the defect candidate, the frequency of generation of the above-described noise component with respect to the interval is obtained, and the frequency threshold is defined based on the generated frequency. The mode 17 of the present invention provides a processing method of the above-described type 丨6, wherein the above-described simulated image is produced in such a manner that the density of the noise components in the image differs depending on the image area. The type 18 of the present invention provides a method of processing 141812.doc 201009328 as described in the above-mentioned Types 14 to 17 of the above-mentioned items, which makes it possible to discriminate between the above-mentioned points of interest as intervals of periodic defect candidates in the above moving direction. When the pitch interval is made, after the step of determining the above, the region of interest including the defect candidate having the pitch interval in the width direction is further defined; and the image from the region of interest is used using the second signal threshold I′ Extracting a detailed defect candidate from the beginning of the image; defining a φ search area centered at a position at which the detailed defect candidate obtained from the extraction in the moving direction is separated from the pitch interval; and using the above-mentioned search area The second signal threshold searches for detailed defect candidates; respectively, evaluates the detailed defect candidates detected by the search, and the attributes of the detailed defect candidates obtained by the extraction; and determines whether the region of interest includes the periodicity in the moving direction according to the evaluation result. Defect candidates. According to a tenth aspect of the present invention, in the processing method of the above aspect 18, the image used for the determination of the periodic defect candidate in the region of interest is an image of the plate after the plate-like body is cut at a fixed size. φ is a processing method according to any one of the above aspects 14 to 19, wherein the defect is searched in the moving direction and the defect candidate having the same position in the width direction is detected, and the interval is determined. The attribute of the detected defect candidate or the similarity between the defect candidate and the specific defect candidate 'The above interval is obtained when at least one of the attribute and the similarity satisfies the set condition. A mode 21 of the present invention provides a defect inspection method characterized in that it detects defects existing in a plate-like body, and irradiates light to the surface of the plate-like body while moving the plate-like body relatively. An image of the plate-like body irradiated with light by 141812.doc 11 201009328 is photographed; and the above-described image of the above-mentioned type I4 to 20 is used for the image obtained by photographing. Further, the present invention provides a method for producing a plate-like body, which is characterized in that it is a plate-like body which is conveyed by a conveying roller as a belt-like continuous body; and the defect inspection using the above-mentioned type 13 The apparatus or the defect inspection method of the above-mentioned type 21, inspecting the above-mentioned plate-shaped body during the moving process; and determining, according to the result of the inspection, a conveying roller which causes a defect in the plate-like body on the moving path of the plate-like body; removal or maintenance Determined to transfer the pro. Further, the present invention 23 provides a method for producing a plate-like body, which is characterized in that it is a plate-like body which is conveyed by a conveying roller as a belt-like continuous body; The defect inspection device or the defect inspection method of the above-described type 21 inspects the plate-like body during the movement, and cuts and takes out the plate-shaped body while avoiding the position in the width direction of the defect determined to have the periodic defect. Further, the mode 24 of the present invention provides a computer executable program and a computer-readable recording medium on which the program is recorded, and the program executes the image data for defect inspection as in any of the above-described types 14 to 20. The treatment method. Further, the present invention provides a processing apparatus according to any one of the above aspects 1 to 1 wherein, in the interval, the interval at which the periodic defect is determined in the moving direction is referred to as a pitch interval. Further, the processing unit further defines a region of interest including a defect having the pitch interval at a position in the width direction, and extracts a detailed defect candidate from an image of the region of interest using the second signal threshold, and specifies In the above-mentioned moving direction, the position of the detailed defect candidate obtained from the extraction is separated from the phase 141812.doc -12-201009328, and the search area is centered on the position of the distance of the circumference of the transport roller of the plate-shaped Zhao. Using the second signal threshold search detailed defect candidate to evaluate the detailed defect candidate detected by the search and the attribute of the detailed defect candidate obtained by the extraction, and determining whether the region of interest is in the moving direction based on the evaluation result. Contains periodic detailed defect candidates. Further, the present invention provides a processing method according to any one of the above aspects 14 to 17, wherein the interval between the above-mentioned attention intervals which is determined to have a periodic defect candidate in the moving direction is referred to as After the spacing interval is performed, after the step of performing the above-described discrimination, the region of interest including the defect candidate having the pitch interval in the width direction is further defined; and the threshold value of the 21st is used, from the image of the region of interest, The detailed defect candidate is extracted from the beginning of the image, and the search area in which the position of the detailed defect candidate obtained from the extraction is separated from the position corresponding to the circumferential length of the transport roller that transports the plate-shaped body is defined in the moving direction. Searching for the detailed defect candidate using the second signal threshold; respectively, evaluating the detailed defect candidate detected by the search, and the attribute of the detailed defect candidate obtained by the extraction; and determining the region of interest based on the evaluation result Whether or not the periodic defect candidate is included in the above moving direction. [Effect of the Invention] The image processing apparatus for defect inspection according to the first aspect of the present invention, the defect inspection apparatus of the type 13, the processing method of the type 14 and the defect inspection method of the type 21, and the type 24 In the program and the recorded coal body, the frequency of occurrence and the frequency threshold are used to determine the presence or absence of the period 141812.doc -13 - 201009328. Further, in the above, the frequency threshold is determined according to the interval of attention. When the two frequency thresholds are set to different values, the interval between the above-mentioned attentions of the frequency threshold of the larger one is smaller than the smaller one. The frequency threshold is set by the manner of the above-mentioned frequency threshold. Therefore, even if the captured image contains noise components, the existence of periodic defects can be discriminated. The other types of processing methods for the image data for defect inspection and the method for processing image data for defect inspection according to the present invention are as follows. First, in the type 2 and the pattern 15, the density of occurrence of the defect candidate is set. In the manner in which the target is generated, the first threshold value is set to 'the threshold value' and the frequency threshold value is changed according to the interval of the target, and the value is also changed according to the value of the target generation density, thereby making it possible to discriminate the periodicity more efficiently. The existence of defects. Even if the density of the defect candidates of the plate-like body varies depending on various conditions in the production of the plate-like body, the fluctuation can be randomly dealt with, and the existence of the periodic defect can be optimally determined. In Type 3 and Type 16, the randomly generated noise component can be overcome, that is, even if there is a noise component, the periodic defect can be easily detected without being affected by the noise component. Further, the frequency threshold can be simply specified based on the resolution frequency and the simulation image obtained analytically. In the pattern 4 and the pattern 17, the periodic defects locally generated in the plate-like body can be easily detected by dividing and processing the image. Further, it is also possible to perform inspection by excluding locally generated noise components. In the type 5, since the unit regions of the same size are formed in consideration of the positional deviation of the defect candidates generated in the image of the inspection object, it is more effective to determine the periodicity in a short time by 141812.doc •14·201009328 The existence of defects. In the state 6, the frequency distribution in the width direction of the defect candidate can be obtained and / can be used. The unevenness of the frequency in the width direction and the patterning of the defects are performed. Therefore, in addition to the speculation that it can be effectively used for the cause of defects, it also contributes to the determination of whether or not the sheet can be stably produced continuously. The newspaper helps in the management of production steps. In i匕, 7 , the plate-shaped body is divided into the plate-shaped body region of the set length, and the image of the region is used as an inspection object of a time series unit, and a plurality of time series units are determined. Information on the frequency of occurrence of defect candidates can be expressed as time series data. Therefore, it is possible to more accurately determine the existence of periodic defects, and in addition to grasp the lack of time series changes.

In the case of trapping, it is helpful to use g@ to determine whether the sheet is the early plate-like body (glass substrate), and the processing in the subsequent steps. In the state 8 state, a plurality of time series unit records generate frequency distributions, so that information on how long the periodic defects continue to be generated can be obtained, which in turn facilitates real-time management in the production steps. In Type 9, in the same chart, 圃 τ τ 复 复 复 复 复 τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ τ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ ’ In the type 10, 'finding the interval from the candidate candidate adjacent to the previous one', the interval between the plurality of defect candidates and the previous plurality of defect candidates is obtained, and the generation frequency is obtained, thereby determining the periodic defect. The presence or absence of the periodic defect can be easily detected even in the case where there are defects caused by a plurality of different generation sources, and the positions where the generations are partially overlapped in the plane. 141812.doc 201009328

The accuracy of defect detection. In the type 25 and the pattern 26, the attention area is specified as a detailed defect candidate, and it is determined whether or not the periodicity factor is included. The presence or absence of the defect can be detected in a short time. (4) The assumed defect is rightly extended to the time required for the detection, and the ancient defect is Can improve

It is true that the periodic defects are detected and the interval between the tween is trapped, so that the reliability of the discrimination can be improved. Further, in the method for producing a plate-like body of the form 22 of the present invention, the time required for the repairing operation in the step of providing the conveying roller can be shortened as compared with the prior art, and the management of the step can be continuously performed. It's easy, so you can stabilize your yield. Further, the transfer roller being used in the production step can be easily managed. Further, the period in which the transfer roller is to be replaced in the future can be estimated in advance. In the method for producing a plate-like body of the type 23 of the present invention, it is possible to effectively cut and take out a strip-shaped continuous body, that is, a plate-like body, even if a conveyance roller or the like has a defect, so that good The rate is stable. [Embodiment] Hereinafter, a processing apparatus and a processing method for image data for defect inspection according to the present invention, a defect inspection apparatus and a defect inspection method using the same, and a use method will be described in detail based on preferred embodiments shown in the drawings. There is a method of manufacturing a plate-like body thereof, and a recording medium readable by a computer that records a program for executing the processing method. The defect inspection device 1 of Fig. 1A is a method for performing the defect inspection method of the present invention. 141812.doc • 16· 201009328 The defect inspection device of the present invention determines whether or not the periodicity of the defect is present. The defect inspection device 丨 mainly has a defect inspection unit 1A, a processing portion 16, and a defect inspection unit 26. As the plate-like body of the present invention, a glass plate G having transparency is used in the following description. However, the plate-shaped body of the present invention is not limited thereto, and for example, a long-length acrylic plate or a long film is also included. Or paper, etc. Further, the glass sheet G described below is cut into a strip-shaped continuous body before a specific size, and the conveyance state will be mainly described. As the type of the mother glass substrate to be produced, for example, G6, G8, G10, and G12, etc., tend to have a larger size. In the present invention, a glass plate cut to a specific size may be used as a target. However, in order to determine the periodicity of the defect candidate, it is preferable to use a strip-shaped continuous body, that is, a long glass plate and a glass plate in a conveyed state. Further, in the following embodiments, the glass plate G to be conveyed is imaged using a stationary light source and a camera. However, the present invention may be in the form of a glass plate G that is stationary while moving the light source and the camera φ. In the present invention, it is only necessary to relatively move the light source and the camera and the glass sheet G. The transport direction described in the following embodiments corresponds to the first moving direction of the present invention, and the width direction corresponds to the second moving direction of the present invention. The defect inspection unit 1 shown in Fig. 1A is disposed on a transport path formed by a plurality of transport rollers 11 having different radii. The glass sheet G is a strip-shaped continuous body taken out from the fusion path at a fixed thickness, and is continuously conveyed and moved on the transport path. Here, 'the driving roller is used as the conveying roller 11 for conveying the glass G. However, 141812.doc 201009328 can be used to drive more than one driven roller between the driving parents, and as the conveying newspaper 11, various types of transportation can be used. Roller. For example, it may be a normal conveying roller which is in contact with the glass G as a whole, such as a cleaning roll, a stationary roll, a belt roll, a coating roll, a coating roll, or the like. Further, it is also possible to transport the glass G in contact with the glass G at intervals in the width direction such as a shoulder roller or a step roller. The defect inspection unit 10 includes a light source 22 that irradiates light onto the surface of the glass sheet G, a camera 14 that images an image of the plate-like body that is irradiated with the light source 22, and a processing unit 16 that discriminates the candidate having periodicity. Further, the defect inspection unit 10 may be connected to a display 18a that displays the inspection result or the like as a soft copy image on the screen, and an output system 18 such as a printer 18b that outputs the inspection result as a hard copy image. Or enter the operating system 20, such as a mouse or a keyboard. The defect inspection unit 10 is a device that inspects a defect candidate in a transmission image by transmitting light transmitted through the glass plate g. Further, the processing unit 16 is connected to the defect inspection unit % that reflects the image, and the light source 22 and the camera 24 are disposed on one side of the glass plate G, and the image of the defect of the glass plate G is reflected by the camera. Image and extract the location of the defect candidate. When the glass sheet G is taken out from the melting furnace at a fixed thickness and continuously conveyed by the conveying roller 11, on the surface of the glass sheet G, as shown in Fig. 1B, the conveying roller 丨i is periodically generated. The resulting small defect c^ produces a somewhat flawed defect on the surface of the glass sheet G. Further, it also occurs in the contaminated area Y which is diffused on the crotch region. Further, in the image read by the defect inspection unit 10, in addition to the above-described respective defects, the dot noise randomly generated by the processing at the time of image reading is also regarded as the defect x. 141812.doc -18-201009328 The defect inspection unit 10 is a rough direction of the width direction of the glass sheet G which is generated on the surface of the glass sheet G to be conveyed and which is orthogonal to the conveyance direction in the image of the glass sheet G. The interval d between the defects d which are periodically generated at the same position is distinguished from the defect X by the difference X. The light source 12 is a light source that emits substantially parallel light, which is a linear light source that emits substantially parallel light having a substantially uniform light intensity in a width direction of the glass sheet G (a direction perpendicular to the plane of the paper of Fig. 1A). The light source may be a light source such as a light source or a light emitting diode (LED), and the kind of light is not particularly limited, and white light is preferably used. The camera 14 is a line sensor type camera which is placed at a position facing the glass plate G with respect to the light source 丨2, and directly reads the transmitted light transmitted through the glass sheet G from the light receiving surface. The type of the line sensor of the camera 14 is not particularly limited. It may be a CCD (Charge Coupled Device) type or a CMOS (Complementary Metal-Oxide-Semiconductor) type. The camera 14 is provided with a plurality of stages in the direction perpendicular to the φ paper surface in Fig., and the same position of the transport direction is photographed, and a plurality of cameras are set such that the field of view in the width direction of the glass sheet G partially overlaps each other. The image signal captured by the camera 14 is transmitted to the processing unit 16. The processing unit 16 is a part of the processing device of the present invention which constitutes the method for processing the image data for defect inspection of the present invention. The processing unit 6 generates a portion of the image data to be inspected by the glass sheet G based on the transmitted image signal, and performs defect inspection using the image data. The image signal transmitted from the camera 14 is subjected to averaging processing of the partially overlapped areas 141812.doc -19· 201009328 as described above to generate image data constituting one image. The above image data is used to determine whether or not the detected defect candidate has periodicity. This determination will be explained below. The light source 22 of the defect inspection unit 26 emits a light source that emits light of the surface of the glass plate g, that is, a light source that emits substantially parallel light, and causes light to enter from a direction inclined with respect to the surface of the glass plate G. Like the light source 12 of the defect inspection unit (7), the light source 22 is a linear light source extending in a direction perpendicular to the plane of the paper of Fig. 8, preferably in the width direction of the glass sheet G (the direction perpendicular to the plane of the sheet of Fig. 1) A substantially parallel light having a substantially uniform light intensity is emitted. The present invention is provided with two light sources. For example, an LED light source or a halogen light source can be used as the light source 22. The type of light is not particularly limited and may be red, blue, white or the like, and is preferably white. The camera 24 is a line sensor type camera that collects reflected light emitted from the surface of the glass sheet G and captures a reflected image, and a camera of the same type as the camera 14 can be used. The camera 24 is disposed on the same side as the light source 22 as viewed from the glass plate. The camera 24 and the light source 22 are disposed in a positional relationship between the upstream side and the downstream side in the transport direction, and the light emission direction of the light source 22 and the field of view direction of the camera 24 are incident on the back surface of the glass plate G. The camera 24 is adjusted in the same way. The image captured by the camera 24 is an image that is illuminated by the light source 22 and reflected on the front surface and the back surface of the glass sheet g. This is an image in which a region of the defect existing inside the glass sheet G is a dark portion. The image first includes an optical path in which the defective region is incident on the surface of the glass sheet G by the direction inclined with respect to the surface of the glass sheet G and reflected by the back surface of the glass sheet G. 141812.doc -20 · 201009328 The real image of the defect. Further, a mirror image of the defect formed by the incident light incident on the surface of the glass sheet G from the direction inclined with respect to the surface of the glass sheet G passing through the defective region in the optical path inside the glass sheet G on the back surface of the glass sheet is formed. . Thus, each time the image data obtained by the camera 24 is linearly read and sent to the processing unit 10 逐 as in the case of the defect inspection unit 1 处理, the processing climbing unit 16 performs defect inspection using the transmitted image data. The reason why the defect is inspected by the defect inspection unit 26 is that the defect is located on the surface of the glass sheet G (the surface on the upper side of the glass sheet G in Fig. 1A) and the back surface based on the positional deviation of the real image of the defect and the mirror image in the conveyance direction. (While the surface on the lower side of the glass sheet G in Fig. 1A and the surface on the side of the transport roller 丨) is specified as a defect. That is, when there is no positional deviation and one image is observed, it is known that the defect is on the back side, and when there is a positional deviation amount and two images are observed, it is understood that the defect is located in the glass plate G or the surface of the glass plate G. The defect inspection device 1 shown in Fig. 1A includes two defect inspection units 丨❹ and 26, but the present invention is not limited thereto, and only one of them may be provided. In the processing unit 16, the image data for defect inspection is processed from the image of the obtained inspection target in the following manner. Fig. 2 is a flow chart showing the flow of a defect inspection method, particularly a method of processing image data for defect inspection. First, the inspection condition of the defect to be detected is set (step sl1). Specifically, 5' is preferably input from the operating system 2 (refer to FIG. Α), and the width of the defect 141812.doc -21 - 201009328 is detected by the input of the operator. The allowable amount, the tolerance of the deviation of the transport direction position from the specific pitch interval when the candidate has a periodic defect candidate, and the extent to which the periodic defect candidate is continuously generated, has periodicity and continuous generation. Information on the extent to which a group of defect candidates are generated, and information on the frequency of occurrence of periodic defect candidates. Then, the cell size is determined based on the set inspection conditions with respect to the image to be inspected for the defect inspection (step S110). As shown in FIG. 3, the unit size is obtained by dividing an image of a search target of a search defect candidate into a plurality of parts in the width direction and the transport direction (moving direction) to form a plurality of unit regions having the same size. The length of the unit area in the width direction and the transport direction. The cell size is determined based on the allowable amount of positional deviation in the width direction and the transport direction set as the inspection conditions. For example, the length in the width direction χ transport direction is determined to be 10 mm x io mm. The reason why the cell regions are formed in this manner is that, based on the interval (separation distance) of the cell regions, the following defect candidates located in the cell region and the transfer direction between the candidate candidates located in the other regions are obtained. _ Separately, as long as the defect candidate is located in one unit area, the position of the defect candidate can be considered to be the same regardless of the position. Even if the position of the periodic defect candidate varies within the allowable range in the width direction and the transport direction, the positional deviation of the defect candidate in the cell region can be absorbed or reduced or eliminated by setting the cell size. The interval between stable candidate candidates. Next, it is decided to check the unit length (step sl2). The so-called inspection unit length 141812.doc -22- 201009328 degree is the length of the image transfer direction of the defect inspection object. By repeating the defect inspection by setting the length of the inspection unit to be fixed, the result of the defect inspection in the time series can be obtained, and the cause of the defect can be estimated. The inspection unit length is determined based on the length of the defect as set as the inspection condition, and whether or not the information is periodically and continuously generated. For example, the length of the glass sheet G to be transported for one hour or one day, or the length of l〇〇m or i〇〇〇m, etc., is determined. Then, the target of the defect candidate produces a value of density (step sl3〇). The defect candidate is a dark portion which is divided into a dark portion in the image when the captured image of the glass plate G to be imaged is binarized. In the present embodiment, the periodicity of the micro-scratch caused by the transfer of the glass sheet G is periodically checked. Therefore, when the density of occurrence of the defect candidate is high, a plurality of spots due to noise components are formed. Dark area. Therefore, it is difficult to accurately determine whether or not the micro-scratch to be inspected periodically reduces the first signal threshold value described below and reduces the generation density of the defect candidate, and it is also impossible to divide the micro-injury to be inspected as a dark portion of the defect candidate. Open situation. Therefore, the target generation density of the dark portion in the image is determined using the information of the frequency of occurrence of the periodic defect candidate set as the inspection condition. π Next, the threshold Hf is determined based on the determined target generation frequency (step S140). The second signal threshold is a threshold value of the image data when the captured image of the glass plate G is binarized. When the value of the image data is lower than the value of the second signal threshold 1418l2.d • 23·201009328, it is divided into a dark portion (the defect candidate processing unit 16 is provided with a reference indicating the relationship between the threshold value and the density of the dark portion). Table, according to the target of the determined density, the main poor day „主4?111_^1_ /,,,, this knowledge table does not appear and determine the threshold signal threshold. Generally speaking, the target density is smaller. Then, the second signal threshold is set to be lower. The reference table is obtained in the defect inspection unit 10 for the captured image of the specific glass plate G in advance, and the relationship between the ith signal threshold and the density of the dark region is obtained and stored in advance. In the memory, the frequency threshold is determined according to the value of the target occurrence density of the defect candidate (step (10)). The frequency threshold is used to discriminate the glass plate in the defect inspection (step S16〇) described below. Whether G has a periodic defect value. The inter-frequency value is used to determine the frequency of periodicity in the histogram obtained by the horizontal axis indicating the interval of the defect candidate and the vertical axis is not relative to the frequency of the interval. For example, when the processing unit 16 creates a histogram as shown in FIG. 4, it is determined whether or not the frequency of the interval B is larger than the frequency value a determined for the interval B. Periodically, when the frequency of generation is higher with respect to the frequency threshold A, it is judged to have periodicity, and the interval B is set as a pitch interval. Whether or not the periodicity is tied to each of the horizontal axes of the histogram The frequency threshold is changed according to the interval of the attention, and the frequency is set to be larger as the interval is larger. Further, it is preferable that the frequency threshold is changed in addition to the interval. It is changed according to the value of the density of the target of the defect candidate. Specifically, it is better to use 141812.doc •24 201009328. The smaller the density becomes, the smaller the value of the sentence between the frequencies is. The setting is set in such a manner that the frequency threshold value is set in this way (the reason is that since the defect candidate includes the defect candidate due to the noise division-α, the defect candidate is set as described above. It is determined to prevent the occurrence of defects caused by the noise of μ, Α > the amount of knives. Figure 5 Α and 5 Β indicates that only the roots are q 八旦. The interval between the candidate candidates in the simulated image produced by the churning component, and the relationship between the frequency of the defect (4) and the frequency of the interval

Chart. Assuming that the noise component 俜醅MA m is randomly generated, the glass component g is assumed to generate a signal component in a region of length 2500 mm, unit scale + 彳π 1 Λ flat size 10 mm x 10 mm. The vertical axes of Figs. 5A and 5B each indicate the number of generations per j m2 (generation frequency). The generation density of the noise component in Fig. 5A is divided into three kinds (50/W, m2, eagle according to Fig. 5A, for any - generation density, the smaller the interval, the higher the frequency of generation) the greater the density of noise components Therefore, the frequency is higher. Therefore, in order to prevent misidentification of the periodicity of the defect candidate due to the noise component, it is preferable to set the frequency threshold to be generated with respect to the vertical axis shown in FIG. 5A (having The margin is higher. For example, 'the value of 1.1 to 2 times the frequency of occurrence of each interval is set as the frequency threshold. Of course, the above value can be set larger according to the inspection condition. Therefore, ::: : The smaller the density of the large and noise components, the lower the above-mentioned frequency. The lower the number is, the lower the value is. The lower the so-called setting, the lower the frequency threshold does not change even if the left-hand spacing is different. For example, the case is as follows: the interval of attention is 2 〇〇 The frequency is greater than the frequency threshold of 500 mm and 1〇〇〇mm, but the interval is 14I812.doc •25- 201009328 500 mm Frequency threshold and 1〇〇〇mm The frequency threshold is equal. In the present invention, when the two frequency thresholds are different, the frequency threshold is set in such a manner that the interval of the frequency threshold of the larger one is smaller than the interval of the frequency threshold of the smaller one. Fig. 5B is a graph showing the frequency of generation of the setting interval MM, 750 mm, and 500 mm with respect to the density of the noise component. According to Fig. 5B, the smaller the generation density of the noise is, the interval is set. The frequency of occurrence of each of the melon melons, mm, and 500 mm becomes smaller. Further, as is known from Fig. 5B, if the density of the noise components reaches a certain level or more, the frequency of occurrence is lowered. That is, the frequency of occurrence has a peak value with respect to the density of the knives. The reason is that the noise component becomes 1 〇〇〇 mm, 75 〇 mm, 500 mm due to the increase in the density of the noise components. Between each set interval, the frequency of occurrence of each set interval of 1 〇〇〇 claw, 750 mm, and 500 mm is lowered. Therefore, the target of the above defect candidate is density to determine whether there is a week or a month. It is defined in such a manner that the interval between the birth points is smaller than the value of the density of the peak positions. In the example shown in FIGS. 5A and 5B, the defect candidates in the simulated image formed based on the noise components are used. The frequency threshold is determined as a result of the interval and the frequency of occurrence of the interval, but the present invention is not limited to the case of using an analog image formed based on the noise component. For example, 'the noise component may be assumed to be randomly in the region. Distribution, the noise component of the region in the width direction at the same position as the defect candidate 'analytically find the frequency of generation of the noise component, and according to the frequency of the generation of 141812.doc -26- 201009328 And determine the frequency threshold. The fact that the generation frequency is obtained analytically means that the generation frequency is calculated using the equation. For example, in the one-dimensional region, the probability pp for finding the η group spacing is expressed by the following equation. At this time, the probability ρρ is calculated by changing nKi to Ν/ρ (the ρ is expressed by the unit unit), and the expected value is added, whereby the frequency of occurrence with respect to the pitch ρ can be obtained. Ρρ^ρΜι-ρΠα Here, the probability that the Ρ is the noise component in one unit, the total number of lanthanum units, and the length of the Ρ is expressed in units. (Nnp) C/C represents a combination of combinations. Further, as another aspect, it is possible to use the defect inspection unit 10 in the glass plate G in which the periodic defect is not caused by the conveyance roller, and to determine the frequency of occurrence of the noise component by actual measurement, and use the The frequency threshold is determined by the measured result. The frequency of defect candidates caused by the noise component is actually uneven in each area of the image, and there is also a case where the noise component is not generated by the same density in the entire region. Therefore, the frequency threshold after the above unevenness can be determined in consideration of the actual measurement using the glass plate G. In this respect, it is preferable to determine the relationship between the interval between the actual measurement results and the frequency of occurrence of the defect candidates in advance, and use the relationship to determine the frequency threshold. Further, even in this case, the frequency threshold is set so as to be substantially larger as the interval is smaller. Fig. 6A is a graph showing the frequency of occurrence of the interval at the time of defect inspection (measured) with respect to the defect inspection unit 1A. The symbols in the graph indicate the results of actual measurement using the defect inspection unit 10. On the other hand, the result obtained by 141812.doc -27-201009328 lacking: the generation density of the candidate (average density of occurrence), which results in the generation of the noise component of the noise component, is represented by the symbol ◊. As can be seen from Fig. 6A, when the interval between the defect candidates is · or less, there is a deviation between the frequency of the actual measurement and the frequency of the simulation. The reason for this is that even if the average density of the noise components in the simulated image is the same as the average density of the actual measurement, the density of the defect candidates in the image of the object to be inspected which is actually measured varies depending on the region. Actually, when the target of the inspection is actually measured, the number of defect candidates is divided and the probability density function of the defect candidate is checked. As shown in Fig. 6B, the probability density function has a distribution. Therefore, in order to conform to the measured probability density function, by making the same probability density function in the simulated image, as shown in Fig. 6C, the symbol ◊ of the simulated image is used to indicate the frequency of generation of the symbol close to the actual measurement. Therefore, it is also possible to determine the relationship between the interval and the generation frequency based on the probability density function used for the generation of the noise component, the analog image obtained by the distribution in accordance with the measured method, and use the relationship to determine the frequency threshold. As described above, it is also possible to perform the actual measurement by using the glass plate G having no periodic flaws, and to prepare the relationship between the interval and the frequency of occurrence of the defect candidate, and use the relationship to determine the frequency threshold. Next, the defect inspection is performed (step S160). In the defect inspection, first, the image of the inspection object sent from the camera 14 is cut out to the area of the inspection unit length and the image of the area of the unit length is examined, as shown in FIG. Regionalization. The zoning of the image is performed based on the cell size determined by the image of the search target candidate 141812.doc -28. 201009328, and a plurality of cell regions having the same cell size are formed. When the plurality of unit regions including the plurality of defect candidates are located at the same position in the width direction, the defect candidates are in the same position in the width direction as described above, and the interval between the defect candidates and the generation frequency are obtained. Then, using the determined signal threshold (threshold Hf threshold), the image of the region of the unit length is binarized, and the plurality of dark regions are extracted as candidates for the defect, and the direction of the image is transferred from the image. The end portion repeats the search along the transport direction and detects the defect candidate. The defect candidate detection is performed in a single 7L size division unit. If there is a defect candidate in the division of the unit size, the width direction of the representative point of the division (the center point of the division, or the moment opening > the apex of the division) The position and the position of the transport direction are stored in a memory (not shown) of the processing unit 16. Further, it is searched for whether or not there is a defect candidate along the transport direction. When the defect candidate is detected, the position in the transport direction of the defect candidate having the same position in the width direction stored in the memory is called, and the position of the transport direction of the found defect candidate and the position of the transferred transport direction are obtained. Differential and set the difference to interval. Further, the count value which is determined in each of the positions in the width direction, which is set in the memory (not shown) of the processing unit 16 and which is determined at each interval, is advanced by one. The position in the transport direction and the position in the width direction are expressed by the value of the representative point of the division divided into unit sizes. In this manner, the search and detection of the defect candidates are performed until the entire image of the inspection target is searched. Finally, using the value of the memory 1418I2.doc -29- 201009328 stored in the memory, the frequency of occurrence of the defect candidates at the positions of the width directions and the other intervals of M and the mother are obtained. Next, in the processing unit 16, a histogram of the defect candidate shown in Fig. 4, the horizontal axis indicating the interval, and the vertical axis indicating the frequency of occurrence is shown in Fig. 4 (step S17). Specifically, the frequency of occurrence of the position in the frequency of the memory and the direction in the width direction stored in the memory is accumulated at each interval, and a histogram indicating the frequency of generation of each interval is created. By summing the frequency of occurrence of each of the intervals noted in the histogram, it is checked whether the frequency of occurrence of the interval of interest is higher than the determined frequency threshold (frequency threshold A in Fig. 4). When the generation frequency is higher than the inter-frequency value, the interval corresponding to the generation frequency is judged to be an interval formed by the periodicity of the defects, and it is judged that the glass sheet G has a periodic defect (step S180). Since the glass sheet G is continuous in the shape of a long strip in the transport direction, the image of the determined unit length is subjected to the above-described defect inspection as an individual unit, and the time series is sequentially The production frequency is produced for each inspection unit length. Of course, the images of the plurality of units produced are respectively used as the object of defect inspection. In this way, the defect inspection of the long glass sheet G conveyed on the conveyance path is performed. As a result of the above defect inspection, a defect is generated at the interval between the judged intervals, and therefore, the information of the pitch interval can be used to infer which roller of the transport roller is caused by a roller having a diameter. In the present embodiment, after step S1 80, it is further possible to infer the occurrence of defects having periodicity in accordance with the flow shown in Fig. 7 141812.doc -30-201009328. First, the frequency distribution 'determined in the width direction determined by the predetermined interval in step S180 is calculated and the feature amount of the distribution is calculated (step S1 81). The characteristic quantity of the distribution includes, for example, the position in the width direction of the maximum generation frequency and the standard deviation of the generation frequency in the frequency distribution in the width direction.

Since the frequency of each of the intervals and the width direction is stored in the memory of the processing unit 16, the interval can be fixed to the predetermined interval, and the frequency distribution of the position in the width direction can be obtained. Three examples of the distribution of the specific interval in the width direction are shown in Figs. 8-8, 8B and 8C. The distribution of the width direction described above is preferably shown on the screen of the display 18a (see Fig. 1A). The example of Fig. 8A produces a pattern in which the frequency is projected at a position in the width direction. The example of Fig. 8B is a generation pattern in which a frequency is generated dominantly in a fixed range in the width direction and a distribution is formed in the range. The example of Fig. 8C is a pattern for generating a frequency in which the frequency is uneven over a wide range in the width direction. The frequency distribution of the position in the width direction is classified into a plurality of generation patterns in addition to the standard deviation (unevenness) of the distribution using the feature amount such as the position in the width direction of the maximum generation frequency. Then, the defect candidate which is determined to have a periodic defect (hereinafter referred to as a candidate candidate having a periodicity) is obtained, and the distribution of the time series of the generation frequency and the generation density is obtained, and the feature quantity of the distribution is calculated (step S182). As described above, the length of the inspection unit is determined as the inspection object of one time series unit, and the defect inspection is performed on a plurality of time series units in sequence, because 141812.doc •31 - 201009328 can produce a periodic defect. The generation density of the candidate (/m2) and the time series distribution of the generation frequency. Fig. 9 is a diagram showing the time in which the generation density (generation frequency) of the specific interval interval, the generation density of all defect candidates, and the frequency of occurrence of the width direction of the specific interval interval (standard deviation) are overwritten in one graph. An example of a sequence distribution.

Further, as the feature quantity of the distribution, the value of the distribution of each time series is set, and the duration of the generation of the enthalpy is determined. Or find the phase relationship of each time series distribution. Or 纟, as the characteristic quantity of the distribution, the standard deviation of each time series distribution is obtained. Further, the range of the value of the vertical axis of the time series distribution of Fig. 9 is different for each case. The above time: The column distribution is preferably displayed on the screen of the display 18a. Further, for the time series distribution (time-sequence data) of the frequency of occurrence of the defect candidate, a plurality of time-series distributions of the rate of occurrence of at least the square of the position between the intervals of the attention and the width direction can be obtained (time series) Data) overwritten in the same chart, and on the screen of the display (10)

Further, a method of expressing the frequency of generation of the defect candidate, a method of expressing the time-series distribution such as the frequency of generation of the trapping candidate, and a time-series representation method of the frequency of occurrence of the candidate complement will be described later. In the 乂 interval, the 'following' is determined with respect to the plurality of pitches determined to have periodicity, and the candidate candidate having the periodicity noted in step S182, the defect interval of the interval interval < the frequency of generation in the width direction Distribution 141812.doc -32- 201009328 Create the spacing interval $4: soil step # Μ > m gate as it is missing, the distribution corresponding to Figure 9, and display on the screen of display 18a. At this time, J = the frequency distribution of the width direction and the time series distribution of S181 and 182. The frequency distribution and the sequence of the width direction of the sigma Each correlation between them (step s! 83). Specifically, the correlation coefficient is obtained for each of the frequency distribution in the width direction and the time series distribution. Then, the feature quantity and the time series of the frequency distribution of the width direction of the other pitch defect candidates are calculated as the feature quantities of the cloth, and are compared with the feature quantities calculated in steps 181 and (8). Further, 'calculating a plurality of feature quantities α (image size, shape, image data value, etc. of the defect candidate) that are determined to have an image of a plurality of periodic defect candidates, and calculating an average value including the feature amount α and The feature amount β (step S184). Poor

The feature quantity α and the feature quantity ^ calculated in this way are compared with the feature quantity calculated in step (4) and 182 - and compared with the feature quantity stored in the library in advance as the past data 'When it is compared (4) The cause of the defect registered in association with the feature amount is called in the allowable range, and is caused to be the cause of the periodic defect (step si85). Furthermore, it is assumed that the other gaps 3 which are evaluated to have higher correlation by the step S183 are caused by the same reason as the inferred cause, for example, due to the plurality of transport rollers 11 in the same period. When the same foreign particle size and the same material are on the surface of the same conveying roller, when two foreign substances of the phase adhere to different positions on the circumference of 141812.d〇c • 33 - 201009328, two defects appear between the candidate candidates. Spacing interval. However, the characteristic quantities of the frequency distribution and the time series distribution in the width direction of the defect candidate are calculated and compared, thereby having the same generated frequency distribution ' and having the same time series distribution. Therefore, it can be inferred that the defect candidates which are evaluated as the other pitch intervals which are highly correlated by the step S183 and the defect candidates which are the target in the step 181 are generated by the same transfer roller. Furthermore, the comparison of the above feature quantities (consistent, inconsistent) can be compared by setting the conditional differences of the parent feature quantities, and the feature quantity can be used _ Mahalanobis space and Mahalanobis The distance can be compared and the comparison can be made by constructing a neural network. Assuming that two pitch intervals are formed between the defect candidates, in the present invention, in addition to finding the interval between adjacent defect candidates (previous defect candidates), the interval between the candidate candidates can be obtained. The interval between the defect candidates adjacent to the adjacent defect candidates (two preceding defect candidates), and the interval between the defect candidates (three preceding defect candidates) adjacent to the two preceding defect candidates ...determine the interval between the defect candidates (N former defect candidates) adjacent to (N-1) (N is an integer of 4 or more) before the _= defect candidate. For example, in the step In the defect inspection in the 16〇, the adjacent defect candidates are sighed as the object to obtain the interval. In step 8181, when two or more intervals are generated, the interval information stored in the memory may be used. Wei ill ', the interval between the adjacent defect candidates (previous defect candidates) and the interval between the candidate candidates adjacent to the plurality of previous defect candidates. 141812.doc -34- 201009328 In the case of the it shape, the upper limit of the interval to be determined is the maximum circumference of the transport roller 丨. In this case, the frequency of occurrence of the interval associated with all combinations of the i to N preceding defect candidates may be determined to be a histogram, and the frequency threshold may be determined using the frequency threshold set separately. There is a periodic defect, and with this configuration, the circumference of the conveying roller U is (10). In the ugly situation, when the interval between the candidate candidates is 3 (10) and the frequency of the generation of the face exceeds the frequency threshold, the frequency of the generation of 1000 mm is added to the sum of the frequencies of j3〇〇mm and 700 mm, so The frequency at which the frequency threshold is set. Thereby, it can be more accurately estimated that the defect candidate system is generated by the same conveying roller. Fig. 10 is a view for explaining a flow of an example of the execution of the defect inspection shown in Fig. 2, and may be carried out as follows. In step s16 of FIG. 2, as described above, in the defect inspection, the image of the inspection object transmitted from the camera 14 and produced is divided into an area of the inspection unit length, and the image of the area of the unit length is as shown in the figure. 3, the image is divided by the unit size (step S16D. Secondly, the image of the area of the unit length is binarized using the determined first signal threshold, and the area on the dark side is used as the defect candidate. The end portion of the image in the transport direction searches for the defect candidate along the transport direction, and extracts the position in the width direction of the detected defect candidate (step S162). At this time, if the defect candidate is detected, the defect is discriminated The attribute of the candidate (step Si63). As the attribute, for example, whether the value of the image signal of the defect candidate is less than a specific value, the area of the image of the defect candidate, or the shape of the defect candidate satisfies the set condition. The defect candidate J41812.doc -35· 201009328 is placed on the back side of the glass plate G (the side of the conveyance roller 11). The defect candidate is located on the back side, or The attribute on the surface can be discriminated using the reflection image obtained by the defect inspection unit 26. That is, the reflection image is irradiated on the back surface by the light irradiated on the back surface of the glass sheet G, and is imaged by the camera 24, so as described above, As described above, when there is a defect on the back surface, there is no deviation between the real image and the mirror image in the reflected image. On the other hand, there are real images and mirror images of the defects existing on the surface, and the positional deviation amount is fixed according to the thickness of the glass plate G. In this respect, it is possible to determine whether or not the defect candidate of the reflected image located at the position corresponding to the defect candidate is located on the back side. The attribute is discriminated using the result of the determination. It is identified by the shape of the image of the corresponding defect candidate obtained from the reflected image. Next, 'the position of the conveyance direction and the position of the width direction of the division to which the defect candidate corresponding to the desired attribute belongs is proposed (step S164), and the division is made. The position of the representative point in the width direction and the position of the transport direction, and the image area of the defect candidate In the memory (not shown) of the processing unit 16, the direction of the transport is further searched for whether or not there is a defect © the desired attribute. In this case, the image area of the defect candidate detected as the desired attribute is already stored in the memory area. The correlation between the size and the correlation coefficient of the shape of the image region is obtained between the image regions of the defect candidates in the memory, and the correlation and the result are evaluated as similarities (step My). When the correlation coefficient is used for evaluation, when the value of the correlation coefficient exceeds a certain value, it is judged that the similarity is high. As the evaluation object of the similarity, the detection is 141812.doc •36-201009328 The candidate candidate and the memory in memory The defect candidate is, for example, a defect candidate generated at a specific interval, or a defect candidate detected and a defect candidate detected and memorized before the detection as a parent group, or a detected defect candidate Defect candidates and defect candidate models of the pre-requisite f. The related objects include the feature amount of the defect candidate, the image data of the defect candidate, or the image feature amount (the feature amount of the shape, etc.) of the defect candidate. ❹ Furthermore, the evaluation of similarity can be evaluated by using the characteristic amount of Mahalanobis space and Mahalanobis distance, in addition to the correlation, or by constructing a neural network. Evaluation. The difference between the position in the transport direction of the defect candidate at the position in the same width direction that is stored in the memory and in the memory is determined with respect to the position of the transport direction of the defect candidate that is determined to have a high degree of similarity. . The count value of the frequency of occurrence of each position and interval indicated in the width direction is advanced by one relative to the interval ' as the difference' (step S be).判断 It is judged whether or not the above defect candidate search and detection have been performed for the entire image of the inspection target (step S167), and if the determination is negative, the process returns to step S162. If the above determination is affirmative, the process proceeds to step S170. When the defect candidate is detected as described above, the condition of the desired defect candidate and the image region of the defect candidate are strictly set, and the defect candidates to be detected can be restricted. Of course, any one of the similarity of the image of the defect candidate and the image area of the defect candidate may be set as the condition.乍 is the step after the step shown in FIG. 2, and the cause of the defect can also be removed according to the flow shown in FIG. 141812.doc -37-201009328. If it is determined in step S180 that there is a candidate candidate having a periodicity, the frequency distribution and the time-series distribution in the width direction of the interval between the candidate candidates are obtained as described above, and the generated frequency distribution is obtained based on the obtained The feature quantity of the time series distribution is extracted to extract features of the pattern, thereby evaluating candidate candidates having periodicity. Alternatively, the defect type of the defect candidate is identified (step S191). The type of the defect refers to the shape of the image of the defect candidate in the reflection image captured by the defect inspection unit %, and the type of the defect determined by using the value of the image data indicating the degree of the light intensity, for example, the spring glass plate G The wound # produced on φ, or the attachment on the surface of the glass plate g, and the like. Then, based on the evaluation result or the recognition result, it is inferred that the cause is that the conveyance roller is defective (step S192). For example, it is inferred that a conveying roller having a circumference conforming to the pitch interval is used as a cause. Further, it is estimated which of the conveyance rollers causes a defect depending on the type of the defect. These inferences can be made by pre-establishing a database that correlates the cause of the defect with the type of defect or pattern creation. ❹ Next, the cause of the removal is performed based on the estimation of the cause (step S193). For example, when the cause of the transfer is specific, the transport path is controlled by moving the specific transport roller from the transport path, and the transport is automatically removed from the transport path, and the transport path is replaced with another transport light. And the drive device operates. Or, you can customize automatic maintenance for a specific transfer. As the automatic maintenance, an example of a protective film for thickening and conveying a light surface can be cited. 141812.doc -38- 201009328 The cause of such a %• is associated with the pattern or the type of defect, and is additionally registered in the above-mentioned database (step si94). Of course, when the reason for the inference is incorrect, the amendment is executed by the operator's input and registered in the library. This database is used for the generation of the original in step S82. Alternatively, the cutting machine may be subjected to the step t of cutting the glass sheet G after the conveyance, and the cutting machine receives the instruction 'to align the glass sheet to a specific size so as to avoid the above-mentioned width direction position 判别 which is determined to have a defect of the periodic defect. Cut and cut out. Further, in the present invention, it is possible to use the data of the frequency distribution of the interval between the above-mentioned defect inspection and the width direction of the defect candidate, and to easily detect the existence of the periodic defect by using the method described below. In other words, the processing unit 16 determines the region of interest including the position where the defect candidate having the pitch interval is located in the width direction. In the attention area, the detailed defect candidate is extracted from the beginning of the image using the second signal threshold value from the image captured by the camera 14. ❹ Searching for the detailed defect candidate using the second signal threshold as the center search area with the position of the detailed defect candidate obtained by the extraction in the transport direction as the center, and evaluating the detailed defect candidate obtained by the search and The similarity of the images between the detailed defect candidates obtained is extracted. Based on the evaluation result of the similarity, it is determined that there is a periodic defect in the transport direction in the region of interest. The second signal threshold may also be set to a lower value than the ith signal threshold determined in step S150 of Fig. 2 . Figure 12 shows an example of the flow of the inspection. First, as shown in FIG. 13A, a region of interest 包含 which is obtained by the defect inspection of the flow shown in FIG. 2 and which is determined to be 141812.doc -39-201009328 has a position in the width direction of the periodicity, and In the region of interest 详细, the detailed defect candidate is detected from the beginning of the image using the second signal threshold (step S195). In Fig. 13A, the defect candidate D丨 is detected as the detected detailed defect candidate. Then, based on the position of the detected defect candidate D] in the transport direction, as shown in FIG. 13B, a search range Α γ of a fixed range is set as a center from a position at which the position is separated from the pitch interval in the transport direction. S196). The detailed defect candidate is extracted using the second signal threshold in the set search area ( (step S197). Next, the detailed defect candidates detected in step S195 and the attributes of the detailed defect candidates are evaluated (step S198). As the attribute, for example, the position at which the defect is generated (surface side or back side) is discriminated. As described above, the discrimination is determined based on the image data supplied to the defect inspection unit 26 of the processing unit 16 using the difference between the defect candidate located on the surface and the defect candidate located on the back surface and the position difference 实 between the real image and the mirror image. Next, when the above attributes are identical, it is determined that the periodic defect exists in the region of the glass sheet G of the inspection object (step S199). In this manner, the region of interest AX is determined using the previously obtained interval interval, and an accurate defect candidate is searched for in the region, and a check for the presence or absence of a periodic defect is performed. Further, the inspection method can be applied to a glass sheet G which is cut into a sheet having a specific size, in addition to the long glass-shaped glass G which can be applied to the conveyance. In particular, in the case of a sheet-like glass sheet, the presence or absence of periodic defects can be individually determined using the spacing interval 1418l2.doc -40·201009328. In this type, the region of interest is set for the interval between the funeral and the inspection, but the present invention (4) is not limited to this. The invention can limit the circumference of the conveying roller, the opening of the sputum to the M, the stagnation of the food, the t-shaped and the attention to the circumference of the special delivery roller after the maintenance, etc. In the present invention, in the case of a shoulder roller or a step roller, the distance between the position in the width direction and the conveyance is light, and the relationship may be the same as the circumference of the conveyance roller. The area of interest is set based on this. In the present embodiment, as shown in Fig. 9, a plurality of times such as the time series distribution of the frequency of the defect candidate + phase, the side + ntt and the m frequency are displayed on the screen of the display screen.

The sequence distribution is overwritten with a graph of L 勹 , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , In the present invention, as shown in Fig. 14A, a time-series distribution (time-series data) of the frequency of occurrence of the defect candidate may be displayed on the screen of the display 18a as a two-dimensional density image of the frequency of occurrence of the density table*. The two-dimensional density image is expressed on one of the axes, for example, the vertical axis represents time, and the other axis, for example, the horizontal axis represents the position in the width direction. The color or density (shading) indicates the attention interval corresponding to the condition (for example, A two-dimensional density image of the frequency of occurrence of defect candidates equivalent to the weight of the grain to be transported. In addition, instead of the dimensional density map I, as shown in FIG. 4B, the two-dimensional seat which is the axis of the square and the axis of the other direction is not the right parent. The three-dimensional graph represented by the height of the direction is displayed on the screen of the display 18a. The occurrence frequency of the table defect candidate is higher in the time and width direction position indicated by the black dot of the two-dimensional density image not shown in FIG. 14A, and it is known that the black dot 141812.doc -41 - 201009328 corresponds to FIG. 14B. The peak of the frequency of the two-dimensional — main, the quasi-chart. In addition, as shown in FIG. 15A + μ σ jin does not change the other axis (horizontal axis) to the pitch ', as in FIG. 14A, the time-series distribution of the frequency of occurrence of the defect candidate is expressed as the frequency at which the frequency is generated. The two-dimensional image is displayed on the display; In the two-dimensional density image, one axis (vertical axis) - time and the other axis (horizontal axis) are used as a pitch, and the color or density (light and dark) indicates the width of the respective inspection conditions. A one-dimensional density image of the frequency of occurrence of a directional position (e.g., a position that is prone to occurrence of a flaw of unknown cause). In addition, when the two-dimensional density is also substituted, as shown in FIG. 15A, the two-dimensional coordinates which are the axes of the time and the other axis are orthogonal to each other. The three-dimensional graph represented by the height is displayed on the screen of the display 18a. As described above, in the present embodiment, the parameters (variables) of the frequency of occurrence of the defect candidate include three items of time, width direction, and pitch. Therefore, as shown in FIGS. MA to 15B, instead of the two-dimensional density image or the three-dimensional graph which is the frequency of occurrence of the inter- and inter-width position, the time-series distribution of the frequency of occurrence of the defect candidate can be expressed as shown in FIG. A three-dimensional graph or a two-dimensional density image in which the frequency of occurrence of a specific unit time is serially represented by (10) and the graph nunc. FIGS. 16A to 16C are axes in which one axis is the width direction and the other axis. A three-dimensional graph that indicates the frequency of generation of defect candidates per unit time corresponding to each inspection condition, and indicates the frequency of occurrence of data per day in a time-series manner. Of course, it is also possible to replace FIG. 16A. To the 16C three-dimensional graph, as shown in Fig. 17A to 141812.doc -42 - 201009328 17C, the vehicle (one vertical axis) is set to the pitch, and the other axis is set to the width direction position, χ _ The color or wave (shading) indicates the bottle of the unit time corresponding to each inspection condition, and the two-dimensional density image of the frequency of the side door is changed and displayed. β υ π - here 'Fig. 16Α to l6C And the nc to nc series time series does not show the frequency data generated every day, but it can also be continuously switched to be displayed as a dynamic picture in the display .... The time interval for generating the frequency data is not The special-purpose system can be longer than - day can be shorter than - day, or can be continuous as a so-called dynamic face. Thus, in this embodiment, the interval between defect candidates can be specified by the defect inspection shown in FIG. And a position in the width direction of the defect candidate having the pitch interval, thereby adding a defect having a periodicity as shown in FIGS. 7, 1 and 10 (10). Further, the cause of the defect can be estimated. The defect inspection method on X can be preferably used for manufacturing a glass plate, etc., that is, using the above defect inspection method, the defect inspection is performed during the conveyance of the plate-like body such as the glass plate g, and according to the inspection The result is estimated on the cause of the transport path of the plate-shaped body. The result of the estimation is preferably displayed on the screen of the display device 18a (see Fig. 1A). Therefore, measures for causing the cause of the transport path are obtained. For example, a situation in which a periodic defect candidate is formed due to foreign matter adhering to the transport roller is configured to separate the transport from the transport path by seven. In this way, the transport roller that causes defects such as scratches on the glass is repaired, or it is replaced with another transport roller. Further, in the step of cutting the glass plate G after the transfer of 1418I2.doc •43·201009328, it is possible to avoid the discrimination. The position of the defect in the width direction of the periodic defect is treated in such a manner that the glass sheet is cut and cut at a specific size. In this type, the defect inspection using the present invention is performed by the removal of the stick, maintenance and Feedback to eliminate defects of the glass is performed by replacement, etc., but the present invention is not limited thereto, and feedback of evaluation of the glass manufacturing environment may be utilized. Specifically, f, in the manufacture of glass 4, there is also a case where the contamination in the body does not periodically adhere to the back surface of the glass, and the liquid such as tin (slag) adhered to the lower surface of the glass from the floating groove. The object in the form of a liquid or a semi-liquid is transferred onto the conveyance roller 11 (see Fig. 1A) and re-transferred onto the glass crucible. At this time, the contaminated area Y shown in Fig. 周期性 is periodically generated on the surface below the glass G. Therefore, the glass manufacturing environment of the float bath or the like can be evaluated based on the periodic defect points (contamination) detected on the surface under the glass G. As described above, by applying the defect inspection method and apparatus of the present invention to the inspection of defects caused by a manufacturing environment such as slag, it is possible to evaluate the manufacturing environment such as glass, and to feed back the evaluation results to the manufacture of glass. ® The above image processing method for defect inspection can be processed on a computer by executing a program. For example, the processing program for the image data for defect inspection according to the present invention has the steps of processing the image data for defect inspection by a computer, specifically, a CPU (central processing unh, central processing unit). The order of the person. Programs containing these sequences can also be constructed as one or more program modules. 141812.doc -44· 201009328 The processing program for image data for defect inspection including the order of execution of the t brain can be memorized in the memory (memory device) of the computer or the server, and can also be memorized in the recording medium. At the time of execution, it is read and executed by the computer (CPU) or other computer from the memory or the recording medium. Therefore, the present invention can also be a computer-readable memory or recording medium for storing a processing method for image data for defect inspection which is useful for causing a computer to execute the image data for defect inspection of the above-described type 14. In the above, the processing device and the processing method for the image data for defect inspection according to the present invention are a defect inspection device and a defect inspection method using the same, a method for manufacturing a plate-shaped body using the same, and a method for executing the processing. Although the computer-readable recording medium of the program has been described in detail, the present invention is not limited to the above-described embodiments or examples, and various modifications and changes can be made without departing from the spirit and scope of the invention. The present invention has been described in detail with reference to the specific embodiments thereof. It is understood that various changes and modifications may be added without departing from the spirit and scope of the invention. φ This application is based on a patent application filed on Jul. 18, 2008 (the patent application No. 2008-187450), the contents of which are hereby incorporated by reference. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1A is a view showing a schematic configuration of a defect inspection device according to an embodiment of the defect inspection device of the present invention; and Fig. 1B is a view of a defect inspection device according to an embodiment of the defect inspection device of the present invention; FIG. 2 is a flow chart showing an example of a flow of 141812.doc-45·201009328, which is an embodiment of the defect inspection method of the present invention; and FIG. 3 is a part of the processing of the defect inspection method of the present invention. 4 is a view for explaining a part of the process of the defect inspection method of the present invention; FIG. 5A is a view for explaining an example of a frequency threshold used in the defect inspection method of the present invention; FIG. 5B is a view of the present invention. FIG. 6A is a view showing another example of the frequency threshold used in the defect inspection method of the present invention; FIG. 6B is a view showing the defect inspection method of the present invention. FIG. 6C is a view showing another example of the frequency threshold value; FIG. 6C is a method for inspecting the defect of the present invention. FIG. 7 is a flow chart showing an example of a flow of another embodiment of the defect inspection method of the present invention; and FIG. 8A is a view showing a width direction obtained by the defect inspection method of the present invention. FIG. 8B is a view showing an example of the frequency distribution of the width direction obtained by the defect inspection method of the present invention; and FIG. 8C is a view showing the width direction obtained by the defect inspection method of the present invention. FIG. 9 is a view showing an example of the time distribution of the time sequence 141812.doc 201009328 obtained by the defect inspection method of the present invention; FIG. 1 is a view showing another embodiment of the defect inspection method of the present invention. FIG. 11 is a flow chart showing an example of a flow of another embodiment of the defect inspection method of the present invention. FIG. 12 is a flow chart showing an example of a flow of another embodiment of the defect inspection method of the present invention. Figure 13A is a diagram illustrating the defect inspection method shown in Figure 12; Figure 13B is shown in Figure 12 FIG. 14A is a view showing a two-dimensional density image of another example of the time series distribution of the frequency of occurrence obtained by the defect inspection method of the present invention; and FIG. 14B is a view showing the defect inspection method according to the present invention. A three-dimensional graph of another example of the time-series distribution of the frequency of occurrence is obtained; FIG. 15A is a two-dimensional density image showing another example of the time-series distribution of the frequency of occurrence obtained by the defect inspection method of the present invention; FIG. 15B is a diagram showing A three-dimensional graph of another example of the time-series distribution of the frequency of occurrence obtained by the defect inspection method of the present invention; FIG. 16A is a three-dimensional graph showing other examples of the frequency of occurrence obtained by the defect inspection method of the present invention; Fig. 16B is a three-dimensional diagram sequentially showing another example of the frequency of occurrence obtained by the defect inspection method of the present invention; Fig. 16C is a view showing other examples of the frequency of occurrence obtained by the defect inspection method according to the present invention. 3D chart; FIG. 17 is a time series representation of defects in accordance with the present invention. Method 141812.doc -47-201009328 Two-dimensional density image of other examples of the frequency of occurrence obtained; FIG. 17B is a two-dimensional density of other examples of the frequency of generation obtained by the defect inspection method according to the present invention. Fig. 17C is a two-dimensional density image showing other examples of the frequency of generation obtained by the defect inspection method of the present invention in a time series manner. [Description of main component symbols] 1 Defect inspection device 10, 26 Defect inspection unit 11 Conveying roller 12, 22 Light source 14, 24 Camera 16 Processing unit 18 Output system 18a Display 18b Printer 20 Input operating system 141812.doc •48-

Claims (1)

201009328 VII. Patent application scope: 1. A processing device is characterized in that it is used to inspect a plate-like body by using an image obtained by moving a plate-like body in a special direction and facing the plate-like body. A processing device for image data for defect inspection having a defect; the processing unit includes a plurality of defect candidates extracted from the image using the signal threshold, and the plurality of candidate candidates are extracted from the moving direction Among the defect candidates, a defect candidate having the same position as the position in the width direction of the normal direction, that is, the moving direction is searched for, and the position of the defect candidate detected by the search in the moving direction of the plate-shaped body is obtained and moved. The interval between the positions of the defect candidates adjacent to the defect candidate detected in the direction in the moving direction is repeated by the above-described processing to obtain a plurality of intervals, and the frequency at which the plurality of intervals are generated is obtained. When the frequency of occurrence of the attention interval exceeds the set frequency threshold, it is determined that the plate-shaped body is in the moving direction Having a periodic defect; the frequency threshold used in the processing unit is defined according to the interval between the above-mentioned points of interest, and when the two frequency thresholds are set to different values, the above-mentioned attention is made to a predetermined larger frequency threshold. The frequency threshold is set such that the interval is smaller than the above-mentioned interval of the frequency threshold of the smaller one. 2. The processing device according to claim 1, wherein the processing unit includes a reference table indicating a relationship between a generation density of the defect candidate and the threshold value of the second signal, so that a density of the defect candidate is set to a target generation density 141812. In the method of doc 201009328 degrees, the first signal threshold is set using the above reference table; the frequency threshold is a value that varies according to the value of the target generation density in addition to the change in the interval of interest. 3. The processing device of claim 1, wherein the frequency value used in the processing unit is specified in such a manner that noise components are randomly distributed in the region, and the positions in the width direction are at the same position. The noise component is used as the defect candidate to analytically determine the frequency of generation of the noise component with respect to the interval, or to use the image of the noise component in the analog image formed by the noise component as a defect The candidate obtains the frequency of generation of the noise component with respect to the interval, and defines the frequency threshold based on the obtained generation frequency. 4. The processing apparatus of claim 3, wherein the ten analog images are created in such a manner that the density of the noise amount in the image differs depending on the image area. 5. The processing item of the request item, wherein the processing unit cuts the search target image for searching for the defect candidate into a plurality of portions in the width direction and the moving direction to form a plurality of units having the same size In the region S, when a plurality of unit regions including a plurality of defect candidates are at the same position in the width direction, the defect candidates are set to have the same position in the width direction, and the interval and the generation frequency are obtained. 6. The processing device according to claim 1, wherein the processing unit further obtains 141812 indicating the generation frequency in the width direction with respect to the above-mentioned attention interval ' of the defect candidate determined to have a periodic defect. .doc 201009328 The width direction of the knife cloth generates a frequency distribution, and the width direction is used to generate the unevenness of the above-mentioned generation frequency in the width direction in the frequency cloth, and the defect generation pattern is classified. 7. The processing apparatus according to claim 1, wherein the plate-shaped system has a long shape continuous in the moving direction; and the processing unit divides the plate-shaped body into a region having a plate-shaped body having a predetermined length, The image of the region is used as a check object for one time series unit, and the above-described discrimination is performed for a plurality of time series units. 8. In the case of claim 7, the processing unit records the generation frequency distribution of the interval defined by the interval and the position in the width direction for the plurality of time series units, and generates according to the record. The frequency distribution defines the frequency of occurrence of the interval and the position in the width direction, and the frequency of occurrence is determined, and the data of the inter-material sequence is represented by the i-generation frequency, whereby the information of the defect generation is displayed on the screen. 9. The processing device of claim 8, wherein the processing unit for the time series data of the generated frequency is changed by at least a plurality of times of the position of the interval and the width direction of the above-mentioned attention The sequence data is overwritten in the same chart and displayed on the screen. !〇. As requested! In the processing device, when the processing unit searches for and detects the defect candidate having the same position in the width direction in the moving direction, the interval between the moving directions is obtained except that the adjacent defect candidate is used as the previous defect candidate. In addition, the interval between the plurality of previous defect candidates in the moving direction is also obtained, and a plurality of intervals are obtained by repeating the above-described processing, and the frequency of generation of the plurality of intervals is obtained, 141812.doc 201009328 * When the frequency of occurrence of the interval between the main eyes exceeds the set frequency threshold, it is determined that the plate-like body has a periodic defect in the moving direction. 11. The processing device of the requester, wherein the interval between the intervals determined to have a periodic defect in the moving direction is referred to as a pitch interval, wherein the processing unit further defines a defect candidate having the pitch interval at the width In the region of interest at the position in the direction, and using the second signal threshold, the detailed defect candidate is extracted from the image of the region of interest from the beginning of the image, and the details obtained from the extraction in the moving direction are specified. The location of the defect candidate is a search area centered on the position of the pitch interval, and in the search area, the detailed defect candidate is searched using the second signal threshold, and the detailed defect candidate detected by the search and the extracted result are respectively evaluated. The attribute of the detailed defect candidate is determined based on the evaluation result whether or not the region of interest includes the periodic detailed defect candidate in the moving direction. 12. The processing device according to claim 1, wherein when the interval between the intervals determined to have a periodic defect in the moving direction is referred to as a pitch interval, the processing unit further defines a defect candidate having the pitch interval as described above. The region of interest at the position in the width direction extracts the detailed defect candidate from the beginning of the image from the image of the region of interest using the second signal threshold, and specifies the detailed defect obtained from the extraction in the moving direction. The position of the candidate is a search area centered on the position corresponding to the distance of the circumference of the transport roller that transports the plate-shaped body, and the detailed defect candidate is searched for using the second signal threshold value in the search area, and the evaluation is performed by 141812'doc 201009328 The attribute of the detailed defect candidate detected and the detailed defect candidate to be extracted is determined based on the evaluation result, and it is determined whether or not the region of interest includes the periodic detailed defect candidate in the moving direction. 13. The processing device according to claim 1, wherein the processing unit searches for and detects a defect candidate having the same position in the width direction in the moving direction, and obtains the interval, thereby evaluating an attribute of the detected defect candidate, or The degree of similarity between the defect candidate and the specific defect candidate is obtained when at least one of the attributes and the similarity satisfies the set condition. A defect inspection apparatus characterized by being inspected for defects existing in a plate-like body, comprising: a light source 'which emits light to a surface of the plate-like body; and a camera having one side together with the light source An image of a plate-like body that is irradiated with light by the light source while moving relative to the plate-like body; and the processing device of claim 1, wherein the processing unit of the processing device uses the second signal threshold Extracting the plurality of defect candidates from the image captured by the camera, and searching for a direction of relative movement with respect to the plate body from the plurality of extracted defect candidates in the moving direction That is, the above-described moving direction is orthogonal to the defect candidate in the same width direction. A method for producing a sheet-like body, which is characterized in that it is a sheet-like body which is conveyed as a belt-shaped continuous body by a feed roller 141812.doc 201009328; and the defect inspection as in claim 14 is used. And detecting, by the device, the plate-shaped body during the moving process; determining, according to the result of the inspection, a conveying roller that causes a defect in the plate-like body on the moving path of the plate-shaped body; removing or maintaining the determined conveying roller. 16. A method of producing a sheet-like body, which is characterized in that a sheet-like body which is conveyed by a conveyance roller as a belt-like continuous body is used; and the defect inspection apparatus of claim 14 is used during the movement The plate-like body is inspected, and the plate-like body is cut and taken out by avoiding the position in the width direction of the defect determined to have the periodic defect. 17. A method of processing a defect inspection image which inspects a defect existing in the plate-like body by using an image obtained by moving a plate-like body relative to each other in a direction of a specific direction. Processing method; extracting a plurality of defect candidates from the captured image using the first signal threshold; searching for a plurality of defect candidates extracted from the specific direction, that is, the moving direction, in the moving direction The defect candidate having the same position in the width direction is determined between the position of the defect candidate detected by the search in the moving direction and the position of the defect candidate adjacent to the defect candidate in the moving direction in the moving direction. The interval is obtained by repeating the above processing to obtain a plurality of intervals; 141812.doc 201009328 Finding the frequency of generation of the plurality of intervals; the frequency at which the interval of attention of the field occurs exceeds the value of the ^ The inter-frequency value is the same as the above-mentioned attention. ● When the values between the two frequencies are different, - the square value • +, through the main object of the interval is less than a predetermined small - as a frequency of the square values of the smell ^ attention interval as described later, setting the process as ... request entry side 17 of the intermediate value ^. ❹Check condition; ... before the above-mentioned discrimination, the step density set in δ and the above-mentioned inspection condition becomes the set target generation two candidates, and the first signal reading value is determined; the density is determined by using the reference table to set the frequency threshold. It is a value that varies depending on the value of the above-mentioned target generation density, which varies depending on the interval noted above. 19_ The processing method of claim 17, wherein the inter-frequency value is defined by the following formula: that is, the image of the noise component of the analog image formed by the noise component is used as a defect candidate, and the relative is obtained. The frequency threshold is determined based on the frequency of generation of the noise component at the interval. 1 &gt; 处理 之 19 processing method </ RTI> wherein the above-mentioned simulated image is produced in such a manner that the density of occurrence in the noise knives in the image differs depending on the image area. 21. The processing method of claim 17, wherein the interval between the above-mentioned attention intervals and the interval having the periodic defect candidate is 141812.doc 201009328 in the moving direction, after the step of performing the above determination Further, a region of interest including a defect having the pitch interval at a position in the width direction is defined; and a second signal threshold is used to extract a detailed defect candidate from an image of the region of interest; a search area centered on a position at which the detailed defect candidate obtained from the extraction is separated from the pitch interval in the moving direction; a detailed defect candidate is searched for using the second signal threshold in the search area; The detailed defect candidate detected and the attribute of the detailed defect candidate obtained by the extraction are determined. Based on the evaluation result, it is determined whether or not the region of interest includes the periodic defect candidate in the moving direction. 22. The processing method of claim 21, wherein the image used in the discrimination of the periodic defect candidates of the region of interest is an image of a plate obtained by cutting the plate-like body at a fixed size. The processing method of claim 17, wherein the interval between the intervals determined to have a periodic defect in the moving direction is referred to as a pitch interval, and after the step of performing the determining, further including including the pitch interval The defect candidate is in the region of interest at the position in the width direction; the second signal threshold is used to extract a detailed defect candidate from the image at the beginning of the image; The detailed defect obtained by the extraction i41812.doc 201009328 is a search area centered at a position corresponding to the distance of the circumference of the transport roller that transports the plate-shaped body; and the detailed defect candidate is searched for using the second signal threshold in the search area. And evaluating the detailed defect candidate detected by the search and the attribute of the detailed defect candidate obtained by the above-mentioned extraction; and determining whether the region of interest includes the periodic detailed defect candidate in the moving direction 根据 based on the evaluation result. 24. In the processing method of claim 17, wherein the defect candidate in the moving direction is detected and detected in the moving direction, and the interval is determined, the attribute of the detected defect candidate or the defect candidate is evaluated. The degree of similarity to the specific defect candidate is obtained when at least one of the attribute and the similarity satisfies the set condition. 25. A method for inspecting a defect, which is characterized in that it detects defects existing in a plate-like body; • irradiates light onto the surface of the plate-like body, and moves the plate-like body relatively while photographing An image of the plate-like body irradiated with light; the method of processing according to claim 17 is carried out using the image obtained by photographing. 26. The method for manufacturing a sheet-like body by the method of manufacturing the sheet-shaped body is characterized in that: the sheet is conveyed and conveyed As a plate-shaped body of the strip-like continuum; using the defect inspection method of claim 25, the plate-shaped body is inspected during the movement; the movement path is determined based on the result of the inspection, and is determined by the plate-shaped body 141812. Doc 201009328 A conveying roller that causes defects in the plate body; removes or maintains the determined conveying roller. 27. A method of producing a sheet-like body, which is characterized in that a sheet-like body which is conveyed by a conveyance roller as a belt-like continuous body is used; and the defect inspection method according to claim 25 is used during the movement The plate-like body is inspected, and the plate-like body is cut and taken out by avoiding the position in the width direction of the defect determined to have the periodic defect. 28. A recording medium 'which is readable by a computer and recorded with a computer executable program'. The program executes the processing method of the image data for defect inspection as in claim 17. 141812.doc