JP2018021873A - Surface inspection device and surface inspection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a surface inspection device for inspecting the presence or absence of unevenness on a coated surface of an inspection object. A surface inspection apparatus (1) includes an inspection pattern forming unit (10) for forming a plurality of inspection patterns (P), and a projection unit (20) for sequentially displaying the plurality of inspection patterns (P) on a display screen (21) and projecting them onto an inspection target. Each time the inspection pattern P is projected, a captured image acquisition unit 30 that acquires a captured image of the inspection object 2 and edge extraction is performed on each of the acquired plurality of captured images, and the extracted edges are An edge-extracted image generation unit 40 that generates an edge-extracted image that is shown based on a difference in brightness value between an edge and a portion that is not an edge, for each captured image; The integrated image generation unit 50 that generates one integrated image in which the brightness values are integrated with each other, and the integrated image is divided into a plurality of sections, and the inspection object 2 is based on the integrated value of the brightness values of each section. Includes a determination unit 60 whether the irregularity of the coating surface. [Selection diagram] Figure 1

Description

  The present invention relates to a surface inspection apparatus that inspects the presence or absence of unevenness on a painted surface of an inspection object, and a surface inspection method that inspects the presence or absence of unevenness on the painted surface of an inspection object.

  For example, depending on the product, the surface of the product may be painted, but in such a painting process, the coating material may remain granular on the surface of the product, or it may not be painted with a uniform thickness, It may cause poor painting. Therefore, conventionally, a technique for inspecting the presence or absence of unevenness on the painted surface has been used (for example, Patent Documents 1-3).

  In the light source device and the evaluation method described in Patent Document 1, the illumination pattern formed in a mosaic shape is reflected two-dimensionally on the painted surface so that it is easy to evaluate the paint dust (orange peel) that is poor in the smoothness of the painted surface. The paint surface is judged to be good or bad depending on how it is reflected.

  In the paint state inspection apparatus and inspection method described in Patent Document 2, in order to evaluate paint sand that is poor in the smoothness of the paint surface, thin (1 mm or less) bright stripes are projected and reflected by the paint sand. By analyzing the amplitude width of the bright fringe waveform for the turbulence that is included, the paint dust is quantitatively evaluated.

  In the surface smoothness automatic inspection apparatus described in Patent Document 3, in order to evaluate paint crumbs that are poor in the smoothness of the paint surface, vertical and horizontal two-dimensional stripes are projected and each area divided by the intersections of the stripes The area of each area is calculated, and the quality of the paint hull is judged based on the variation in the area of each area due to the influence of the paint hull.

JP 2012-032174 A JP 2009-204473 A JP-A-1-213509

  In the technique described in Patent Document 1, since the person determines the lighting pattern reflected on the paint surface, the state of the paint scum cannot be inspected quantitatively. May occur.

  Further, in the technique described in Patent Document 2, since a thin (1 mm or less) bright stripe is projected, only the projected portion is evaluated, and it is not easy to inspect the entire painted surface at once. Absent. In addition, depending on the actual paint hull situation, bright stripes may not be reflected properly, noise may be generated due to the influence of paint hulls, and the stripes may be divided into multiple parts, which may prevent accurate waveform analysis. .

  Furthermore, in the technique described in Patent Document 3, it is necessary to project vertical and horizontal two-dimensional stripes and calculate the area of each area divided by the intersections of the stripes. There is a possibility that the fringes are not properly reflected, and noise is generated due to the influence of the coating dust, so that it is difficult to extract the intersections of the fringes and the inspection cannot be performed accurately.

  Therefore, there is a need for a surface inspection apparatus that can inspect the presence or absence of unevenness on the painted surface of the inspection object and a surface inspection method that can inspect the presence or absence of unevenness on the painted surface of the inspection object.

  A characteristic configuration of the surface inspection apparatus according to the present invention is that a first pattern having a lightness of a predetermined level or more and a second pattern darker than the lightness of the first pattern are alternately arranged, and the first pattern and the second pattern A test pattern production unit for producing a plurality of test patterns that are sequentially shifted by a predetermined amount of deviation along the alternately arranged directions, and a plurality of the test patterns produced by the test pattern production unit are displayed on a display screen. A projection unit that sequentially displays the projected image on the inspection target, a captured image acquisition unit that acquires a captured image obtained by capturing the inspection target each time the inspection pattern is projected, and the captured image acquisition unit. Edge extraction is performed on each of the plurality of captured images, and the extracted edges are indicated based on a difference in luminance value between the edge and the non-edge portion. An edge-extracted image generating unit that generates an image for each captured image, and an integrated image that generates a single integrated image in which the luminance values are integrated at the same position in the inspection object across the plurality of edge-extracted images A generating unit; and a determination unit that divides the integrated image into a plurality of sections, and determines the presence or absence of unevenness on the painted surface of the inspection object based on the integrated value of the luminance value for each of the divided sections. It is in the point to have.

  With such a characteristic configuration, the noise at the boundary between the first pattern and the second pattern due to the influence of the coating dust is reduced in the portion where the coating of the inspection target is performed satisfactorily, and the captured image Since the number of edges extracted from is small, the luminance value of the edge included in the edge extracted image can be reduced. On the other hand, the portion with poor coating is included in the edge extraction image because the noise at the boundary between the first pattern and the second pattern due to the influence of the coating dust increases and the number of edges extracted from the captured image increases. The brightness value of the edge to be increased can be increased. In addition, by generating a single integrated image obtained by integrating the luminance values from the plurality of edge extraction images generated in this way, and determining the presence or absence of irregularities on the painted surface of the inspection object based on the integrated image, It becomes possible to two-dimensionally evaluate the painted surface of the object. Further, it is possible to inspect the unevenness (scratch defect) of the painted surface quantitatively from the result of the luminance value integration by the integrated image. Furthermore, by setting a predetermined threshold value for the luminance value of the edge included in the captured image and binarizing to generate an edge extraction image, it is possible to determine whether the paint scrap is acceptable over the entire painted surface. It becomes. Furthermore, it is also possible to identify a poorly painted part on the entire surface of the inspection object.

  Further, it is preferable that the determination unit determines based on an average value of the luminance values for each section.

  With such a configuration, it is possible to accurately inspect for the presence or absence of unevenness on the painted surface of the object to be inspected by comparing the average value of the luminance values with, for example, a predetermined threshold value.

  Further, it is preferable that the determination unit determines based on a standard deviation of the luminance value for each section.

  With such a configuration, by comparing the standard deviation of the luminance value with, for example, a predetermined threshold, it is possible to accurately inspect for the presence or absence of unevenness on the painted surface of the inspection object.

  In addition, the characteristic configuration of the surface inspection method according to the present invention is that a first pattern having a lightness of a predetermined level or more and a second pattern darker than the lightness of the first pattern are alternately arranged, and the first pattern and the second pattern An inspection pattern production step for producing a plurality of inspection patterns that are sequentially shifted by a preset amount of deviation along the alternately arranged direction, and a plurality of the inspection patterns produced by the inspection pattern production step. A projecting step of sequentially displaying on the display screen and projecting onto the inspection object; a captured image acquiring step of acquiring a captured image of the inspection object each time the inspection pattern is projected; and the captured image acquiring step Edge extraction is performed for each of the plurality of captured images acquired by the step, and the extracted edge is divided between the edge and the non-edge portion. An edge extraction image generation step for generating an edge extraction image shown on the basis of the difference in degree value for each of the captured images, and integrating the luminance value for each same position in the inspection object across the plurality of edge extraction images An integrated image generation step for generating one integrated image, and dividing the integrated image into a plurality of sections, and based on the integrated value of the luminance value for each of the sections, the painted surface of the inspection object And a determination step for determining the presence or absence of unevenness.

  Compared with the surface inspection apparatus described above, such a surface inspection method has no substantial difference in characteristic configuration, and the above-described effects can be obtained.

It is a schematic diagram which shows the structure of a surface inspection apparatus. It is a figure which shows a test | inspection pattern. It is an example of a captured image. It is an example of an edge extraction image. It is an example of an integration image. It is the figure which showed an example of the calculation result of the average value of the brightness | luminance for every division which divided the integral image. It is explanatory drawing of the determination by the determination part. It is a flowchart which concerns on determination.

  The surface inspection apparatus according to the present invention is configured to have a function of inspecting the unevenness of the painted surface of the object to be inspected, specifically, the presence or absence of a coating hull called orange peel. Hereinafter, the surface inspection apparatus 1 of the present embodiment will be described.

  FIG. 1 is a diagram schematically showing the configuration of the surface inspection apparatus 1 of the present embodiment. As shown in FIG. 1, the surface inspection apparatus 1 includes an inspection pattern creation unit 10, a projection unit 20, a captured image acquisition unit 30, an edge extraction image generation unit 40, an integrated image generation unit 50, and a determination unit 60. Is done. In particular, each functional unit of the inspection pattern creation unit 10, the edge extraction image generation unit 40, the integrated image generation unit 50, and the determination unit 60 is a hardware that uses a CPU as a core member in order to perform processing related to the inspection for the presence or absence of coating dust. Hardware and / or software.

  The inspection pattern production unit 10 produces an inspection pattern P in which the first pattern F having a lightness of a predetermined level or more and the second pattern S darker than the lightness of the first pattern F are alternately arranged. The “first pattern F having a brightness of a predetermined level or higher” is a display object that is displayed with a brightness of a predetermined level or higher when the test pattern P is displayed on the display screen 21 of the projection unit 20 described later. For example, a display object made of white corresponds. On the other hand, “the second pattern S darker than the brightness of the first pattern F” means that the display object is displayed with a brightness darker than the brightness of the first pattern F when the inspection pattern P is displayed on the display screen 21. For example, a display object made of black corresponds. The inspection pattern P is configured by arranging such first patterns F and second patterns S alternately along a predetermined direction. In the present embodiment, as shown in FIG. 1, the first pattern F and the second pattern S are alternately arranged along the vertical direction of the display screen 21.

  In addition, the inspection pattern production unit 10 produces a plurality of inspection patterns P that are sequentially shifted by a preset amount of deviation along the direction in which the first patterns F and the second patterns S are alternately arranged. FIG. 2 shows a plurality of inspection patterns P produced by the inspection pattern production unit 10. The “direction in which the first pattern F and the second pattern S are alternately arranged” is a direction in which the first pattern F, the second pattern S, the first pattern F, and the second pattern S in FIG. Specifically, the vertical direction of the display screen 21 corresponds. The “deviation amount” is set in advance in the inspection pattern preparation unit 10, and as shown in FIG. 2A, for example, a vertical direction composed of one first pattern F and one second pattern S. If the length is one cycle, the length in the vertical direction (pitch in the display screen 21) is preferably divided into four (the pitch is preferably 2 dots on the image data). In this case, as shown in FIGS. 2A to 2D, four test patterns P are produced in which the first pattern F and the second pattern S are sequentially shifted by ¼ period. The

  In such a direction that the first pattern F having a lightness of a predetermined level or more and the second pattern S darker than the lightness of the first pattern F are alternately arranged, and the first pattern F and the second pattern S are alternately arranged. A process of producing a plurality of inspection patterns P that are sequentially shifted by a preset amount of deviation along the surface is referred to as an inspection pattern preparation step in the surface inspection method.

  Returning to FIG. 1, the projection unit 20 sequentially displays the plurality of inspection patterns P produced by the inspection pattern production unit 10 on the display screen 21 and projects them onto the inspection object 2. The projection unit 20 includes a display unit such as a liquid crystal display or an organic EL display, for example, and displays a plurality of inspection patterns P produced by the inspection pattern production unit 10 on the display screen 21 while sequentially changing them. On the other hand, the inspection object 2 is installed at a position away from the projection unit 20 by a predetermined distance. Thereby, the inspection pattern P displayed on the display screen 21 can be projected (irradiated) onto the inspection object 2 at a predetermined incident angle θ1.

  The process of sequentially displaying a plurality of inspection patterns P prepared by such an inspection pattern preparation step on the display screen 21 and projecting them onto the inspection object 2 is referred to as a projection step in the surface inspection method.

  The captured image acquisition unit 30 acquires a captured image obtained by capturing the inspection object 2 every time the inspection pattern P is projected. As described above, a plurality of inspection patterns P are produced, and the projection unit 20 displays the plurality of inspection patterns P while sequentially changing them. As the inspection pattern P is changed, the captured image acquisition unit 30 acquires a captured image obtained by capturing the inspection object 2 on which the inspection pattern P is projected. FIG. 3 shows an example of a captured image acquired by the captured image acquisition unit 30. The captured image acquisition unit 30 can be configured using an imaging device such as a CMOS image sensor or a CCD image sensor. Note that the captured image acquisition unit 30 may be configured to acquire a captured image obtained by capturing a regular reflection of the inspection pattern P projected onto the inspection object 2. In other words, the reflection angle θ2 may be configured to be equal to the incident angle θ1. Note that θ1 and θ2 are preferably about 30 degrees.

  The process of acquiring a captured image obtained by imaging the inspection object 2 each time such an inspection pattern P is projected is referred to as a captured image acquisition step in the surface inspection method.

  Returning to FIG. 1, the edge-extracted image generation unit 40 performs edge extraction for each of the plurality of captured images acquired by the captured image acquisition unit 30, and uses the extracted edges as the luminance values of the edges and non-edge portions. An edge extraction image shown based on the difference is generated for each captured image. As shown in FIG. 3, the captured image corresponds to a case where the first pattern F and the second pattern S of the inspection pattern P are reflected by the inspection object 2. The edge extraction image generation unit 40 corresponds to a boundary portion between the color based on the first pattern F and the color based on the second pattern S in the captured image for each of the plurality of captured images acquired by the captured image acquisition unit 30. Edge extraction (detection) is performed.

  Here, in the edge extracted image, as shown in FIG. 4, the edge portion is shown in a bright color (for example, white) and the other portion that is not an edge is emphasized so that the extracted edge is emphasized. It is shown in a darker color (for example, black) than the color indicating the edge portion. Therefore, in the edge extracted image, the extracted edge and the non-edge portion are indicated by the difference in luminance value. The edge extraction image generation unit 40 generates an edge extraction image for each captured image, as shown in FIGS. That is, when there are n photographed images, n edge extracted images are also generated.

  Such edge extraction can be performed using, for example, a well-known Canny filter, and the edge extraction image classifies the intensity of the extracted edge by a predetermined threshold, and a portion having an intensity equal to or higher than the threshold. It is preferable to generate 1 with the luminance of 1 being 0 and the luminance of the portion with an intensity less than the threshold being 0.

  Edge extraction is performed for each of a plurality of captured images acquired by such a captured image acquisition step, and an edge extracted image indicating the extracted edges based on a difference in luminance value between the edge and a non-edge portion is obtained. The step of generating each captured image is referred to as an edge extraction image generation step in the surface inspection method.

  Returning to FIG. 1, the integrated image generation unit 50 generates one integrated image in which the luminance values are integrated for each same position in the inspection object 2 over a plurality of edge extracted images. The plurality of edge extraction images are edge extraction images generated for each captured image by the edge extraction image generation unit 40. Here, the positional relationship among the projection unit 20, the captured image acquisition unit 30, and the inspection object 2 is such that a plurality of inspection patterns P prepared by the inspection pattern preparation unit 10 are projected onto the inspection object 2, and the captured image acquisition unit It does not change until 30 acquires all the picked-up images according to the some test | inspection pattern P. FIG. For this reason, “the same position in the inspection object 2” corresponds to the same position in the edge extraction image.

  Further, as described above, in the edge extraction image, the edge portion is shown in a bright color (for example, white), and the other portion that is not the edge is a darker color (for example, black) than the color that indicates the edge portion. Indicated by Therefore, the integrated image generation unit 50 integrates the luminance value for each of the same positions over the plurality of edge extracted images, and one integrated image showing the integrated result at a position corresponding to the position of the edge extracted image. Is generated. An example of such an integrated image is shown in FIG. Note that, for example, a Gaussian filter can be applied so that an edge is not generated due to unevenness or the like in the integrated image and a missing portion is not generated.

  The step of generating one integrated image obtained by integrating the luminance values at the same position in the inspection object 2 over the plurality of edge extracted images is referred to as an integrated image generating step in the surface inspection method.

  Here, it is possible to inspect for unevenness of the painted surface using the integrated image shown in FIG. 5, but the luminance value of the integrated image is uneven, and therefore a predetermined threshold value is used. Even accurate inspection is not easy. Therefore, the determination unit 60 divides the integrated image into a plurality of sections, and determines the presence / absence of unevenness on the painted surface of the inspection object 2 based on the integrated value of the luminance values of the divided sections. To divide the accumulated image into a plurality of sections means to divide the accumulated image into sections of a predetermined size. In the present embodiment, the integrated image is divided into a grid pattern. The size of the grid-like section is preferably a section having a pixel resolution of 10 × 10 dots on the screen data or a size of 1 mm × 1 mm on the surface of the inspection object 2.

  In this embodiment, the determination part 60 determines based on the average value of the luminance value for every division. That is, the determination unit 60 calculates the average value of the luminance in the section for each section divided as shown in FIG. Further, the determination unit 60 plots the calculated result on a map as shown in FIG. 7, for example. In FIG. 7, the vertical axis represents the average luminance corresponding to the above calculation result, and the horizontal axis represents the section. In addition, a threshold is provided for a predetermined luminance value.

  The determination unit 60 plots the respective calculated results. When the average brightness of the target section is lower than the threshold value, the determination unit 60 determines that the painted surface of the section has no unevenness, and averages the target section. If the luminance exceeds the threshold value, it is determined that the painted surface of the section has irregularities. Further, even when the result (average luminance) calculated as in section i or section j in FIG. 7 exceeds (exceeds) the threshold value, for example, the section determined to be uneven with respect to the total number of sections. If the ratio of the number is less than or equal to a predetermined value, the overall evaluation of the inspection object 2 may be evaluated as “no problem” in consideration of misdetection, variation, and the like.

  The step of dividing such an integrated image into a plurality of sections and determining the presence / absence of unevenness of the painted surface of the inspection object 2 based on the integrated value of the luminance values for each of the divided sections is a determination step in the surface inspection method. It is called.

  Next, processing by the surface inspection apparatus 1 and a surface inspection method will be described using the flowchart of FIG. First, the inspection object 2 is placed on the inspection table of the surface inspection apparatus 1 so that the surface inspection apparatus 1 can inspect whether the coating surface of the inspection object 2 is uneven (step # 1). Next, the inspection pattern production unit 10 produces a plurality of inspection patterns P in which the first patterns F and the second patterns S are alternately arranged (step # 2).

  The projection unit 20 displays the inspection pattern P on the display screen 21 and projects the inspection pattern P onto the inspection object 2 (step # 3). The captured image acquisition unit 30 acquires a captured image obtained by capturing the inspection object 2 on which the inspection pattern P is projected (step # 4). An inspection pattern P whose position is shifted from the first pattern F and the second pattern S by a preset shift amount is displayed on the display screen 21 along the direction in which the first pattern F and the second pattern S are alternately arranged. (Step # 5).

  A series of processing from step # 3 to step # 5 is performed until the inspection pattern P shifted by one cycle is displayed with respect to the inspection pattern P displayed first (step # 6: No). When the inspection pattern P shifted by one cycle is displayed with respect to the inspection pattern P displayed first (step # 6: Yes), for each of the plurality of captured images acquired by the captured image acquisition unit 30, The edge extraction image generation unit 40 performs edge extraction, and generates an edge extraction image for each captured image (step # 7).

  The integrated image generation unit 50 generates one integrated image obtained by integrating luminance values using a plurality of edge extracted images (step # 8). The determination unit 60 divides the accumulated image into predetermined sections and calculates an average value of luminance values for each section (step # 9). Furthermore, the determination unit 60 compares each calculated average value with a preset threshold value. If the calculated average value is less than or equal to the threshold value in all sections in the integrated image (step # 10: Yes), the determination unit 60 determines that there is no unevenness on the painted surface of the inspection object 2 (step # 11). The process ends. On the other hand, if any one of the sections in the integrated image has a calculated average value that exceeds the threshold value (step # 10: No), the determination unit 60 determines that the painted surface of the inspection object 2 is uneven. (Step # 12), the process ends.

  The surface inspection apparatus 1 can quantitatively inspect the unevenness of the painted surface of the inspection object 2 by such processing. Further, it is possible not only to inspect the entire painted surface of the inspection object 2, but also to detect a portion having a poor coating state on the painted surface of the inspection object 2. In addition, since the disturbance of the coating surface of the inspection object 2 is extracted as an edge as it is and an integrated image is generated, it is possible to perform inspection with good reproducibility regardless of the shape of the inspection object 2.

[Other Embodiments]
In the above embodiment, the inspection pattern preparation unit 10 has been described as preparing four inspection patterns P in which the first pattern F and the second pattern S are sequentially shifted by a quarter period. The four periods are examples, and can be configured with other numerical values.

  Moreover, although the test pattern production | generation part 10 demonstrated as producing the some test pattern P until the 1st pattern F and the 2nd pattern S shift | deviate 1 period, the some test pattern P which has shifted | deviated 1 period or more is demonstrated. It can also be configured to be manufactured.

  In the above embodiment, the determination unit 60 has been described as determining based on the average value of the luminance values for each section. However, the determination unit 60 calculates the average value of the luminance values while moving the predetermined section. It is also possible to make a determination based on the above. Furthermore, the determination unit 60 may be configured to perform determination based on the standard deviation (luminance variation) of the luminance value for each section.

  INDUSTRIAL APPLICABILITY The present invention can be used in a surface inspection apparatus that inspects the presence or absence of unevenness on the painted surface of an inspection object and a surface inspection method that inspects the presence or absence of unevenness on the painted surface of an inspection object.

1: Surface inspection device 2: Inspection object 20: Projection unit 21: Display screen 30: Captured image acquisition unit 40: Edge extraction image generation unit 50: Integrated image generation unit 60: Determination unit F: First pattern P: Inspection pattern S: 2nd pattern 10: Inspection pattern production part

Claims (4)

A first pattern having a lightness of a predetermined level or more and a second pattern darker than the lightness of the first pattern are alternately arranged, and the first pattern and the second pattern are set in advance along the alternating direction. An inspection pattern production unit for producing a plurality of inspection patterns that sequentially shift by the amount of deviation,
A projection unit that sequentially displays a plurality of the inspection patterns produced by the inspection pattern production unit on a display screen and projects them onto an inspection object;
A captured image acquisition unit that acquires a captured image obtained by capturing the inspection object each time the inspection pattern is projected;
Edge extraction is performed for each of a plurality of captured images acquired by the captured image acquisition unit, and the extracted edges are indicated based on a difference in luminance value between the edge and the non-edge portion. An edge-extracted image generation unit that generates each captured image;
An integrated image generating unit that generates one integrated image obtained by integrating the luminance values for each same position in the inspection object across a plurality of the edge extracted images;
A determination unit that divides the integrated image into a plurality of sections, and determines the presence or absence of unevenness on the painted surface of the inspection object based on the integrated value of the luminance values for each of the divided sections;
A surface inspection apparatus comprising:   The surface inspection apparatus according to claim 1, wherein the determination unit determines based on an average value of the luminance values for each of the sections.   The surface inspection apparatus according to claim 1, wherein the determination unit determines based on a standard deviation of the luminance value for each of the sections. A first pattern having a lightness of a predetermined level or more and a second pattern darker than the lightness of the first pattern are alternately arranged, and the first pattern and the second pattern are set in advance along the alternating direction. An inspection pattern production step for producing a plurality of inspection patterns that sequentially shift by the amount of deviation,
A projecting step of sequentially displaying a plurality of the inspection patterns produced by the inspection pattern production step on a display screen and projecting the same onto an inspection object;
Each time the inspection pattern is projected, a captured image acquisition step of acquiring a captured image obtained by capturing the inspection object;
Edge extraction is performed on each of the plurality of captured images acquired by the captured image acquisition step, and the extracted edge is indicated based on a difference in luminance value between the edge and the non-edge portion. An edge extraction image generation step for each captured image; and
An integrated image generating step for generating one integrated image obtained by integrating the luminance values at the same position in the inspection object across a plurality of the edge extracted images;
A step of dividing the integrated image into a plurality of sections, and determining the presence / absence of unevenness of the painted surface of the inspection object based on the integrated value of the luminance values for each of the divided sections;
A surface inspection method comprising:

Images