JP2002032765A - Image recognizing device and image forming device provided with the image recognizing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image recognizing device capable of reducing even an error of top/bottom recognition while shortening the time taken for top/ bottom recognition processing. SOLUTION: The direction of an image with respect to an original on the 1st page is recognized from the image data of the read original, and reliability P1 to the result is calculated. In the case a relation P1>K holds from the magnitude relation between the P1 and a prescribed value K ('Y' in S13), the image direction recognized on the original on the 1st page is decided as the image direction of all of the read originals (S14). In the case of P1<=K, a return is made to a main routine through S17 and S18. However, when a routine is circulated and in the case of Pn>K on the original on the n-th page in S13 ('Y' in S13), an image direction recognized on the n-th page is defined as the image direction of all of the originals (S14).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image recognition apparatus and an image forming apparatus using the image recognition apparatus, and more particularly to an improvement in a technique for recognizing an image orientation of a document.

[0002]

2. Description of the Related Art In the field of copiers, especially digital copiers, for example, when copying is performed using a binding margin mode for creating a binding margin on the left or right side of a document image, even if a user transfers an original bundle to an ADF ( When the original is set in the automatic document feeder, the orientation of the scanned document image is recognized so that the binding margin is created at the desired position, even if the orientation is wrong. 2. Description of the Related Art Development of a technique of so-called upside-down recognition processing for rotating data by a necessary amount has been conventionally promoted.

[0003] As a conventional top / bottom recognition processing method, for example, when a plurality of originals are read in succession, image data of all the originals is temporarily stored in an image memory, and then the top / bottom recognition is performed for each original. A method of performing image rotation processing on image data of each document based on the result (first related art) or a method of performing image rotation processing on image data of the page while performing top and bottom recognition for each page (second conventional technique) Technology).

However, in the first prior art, since the copying operation is not started until all the originals have been recognized, the start of copying is delayed. on the other hand,
In the second conventional technique, a processing time is required each time to perform the recognition processing for each page, and the time required to copy a document image of one page becomes longer. In both cases, the copy productivity decreases.

Therefore, it is generally considered that each original in a bundle of originals set by the user on the ADF is often in the same direction. Therefore, only the top and bottom of the original image of the first page to be read is read. Japanese Patent Application Laid-Open No. 8-274980 discloses a technique for recognizing and determining that the original images on the second and subsequent pages are in the same direction as the first page (third conventional technique). According to the third related art, since the top and bottom recognition processing is performed only on the first page of the original, compared to the first and second related arts, the time required for the processing is shorter. In addition, since the copying operation can be started immediately after the end of the processing, copy productivity is also improved.

[0006]

However, according to the third prior art, only top / bottom recognition is performed for the original of the first page. Therefore, when the first page is a cover,
Since the text image is surrounded by a frame or contains a chart, it is highly likely that the top and bottom will be misidentified.
For example, when the binding margin mode is used, there is a risk that binding margins may be formed on all the copied sheets on the side opposite to the desired side, and the risk is very large.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and has an image recognition apparatus and an image recognition apparatus capable of reducing the time required for the top and bottom recognition processing and reducing errors in the top and bottom recognition. It is an object of the present invention to provide an image forming apparatus.

[0008]

SUMMARY OF THE INVENTION In order to achieve the above object, the present invention relates to an image recognizing apparatus for receiving image data of a plurality of originals and recognizing the orientation of an image. Recognition means for recognizing the orientation of the image of the original document based on the above, calculation means for calculating the reliability of the recognition result of the recognition means, and, if the calculated reliability is equal to or less than a predetermined value, the reliability The control means controls the recognition means so that the orientation of the image of the document is recognized based on another one page of image data until the calculated reliability becomes greater than the predetermined value. When the value exceeds a predetermined value, a determination unit is provided for determining the orientation of the image recognized with respect to the original as the orientation of the images of all the accepted originals.

Further, the image data of one predetermined page is one page of image data of a document received first. Further, the control means is configured to cause the image data of one page to be recognized in the order of reception of the image data until the reliability becomes larger than the predetermined value.

[0010] Further, when the reliability is equal to or less than the predetermined value for all the originals, the determining means determines a predetermined orientation as the image orientation of all the originals. I do. Further, the determining means refers to the orientation of the image recognized for each document, and when it is determined that the documents recognized as having the same orientation are continuously received for a predetermined number of pages, Even if the reliability does not exceed the predetermined value, the orientation is determined as the orientation of the image of all the originals.

Further, the present invention relates to an image forming apparatus,
The image recognition apparatus, image data rotation means for performing rotation processing on the image data received based on the determination result of the determination means of the image recognition apparatus, and image formation based on the image data subjected to the rotation processing Image forming means. Further, the image forming means starts the image forming operation when the reliability calculated by the calculating means of the image recognition device exceeds a predetermined value.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an image recognition apparatus according to the present invention will be referred to as a digital copier (hereinafter simply referred to as "copier").
A description will be given of a case where the present invention is applied. (1) Overall Configuration of Copying Machine FIG. 1 is a diagram showing an overall configuration of a copying machine according to the present embodiment.

As shown in FIG. 1, the copying machine includes an automatic document feeder 10, an image reading section 30, a printer section 50,
And a paper feeding unit 70. The automatic document feeder 10
This is a known device that automatically conveys a document to the image reading unit 30. Documents set on a document feed tray 11 are separated one by one by a feed roller 12 and a separating roller 13 and sent downward. The document is conveyed to a document reading position on the platen glass 31 by the conveyance belt 14.

The document conveyed to the document reading position is scanned by the scanner 32 of the image reading section 30 and then sent again by the conveyance belt 14 to the right in FIG. It is discharged onto the tray 16.
The image reading unit 30 optically reads an image of a document conveyed to the document reading position on the platen glass 31, and includes a scanner 32 and a CCD image sensor (hereinafter, referred to as a CCD image sensor).
38, etc.).

The scanner 32 is provided with an exposure lamp 33 and a mirror 34 for changing the optical path of light reflected from the original by irradiation of the exposure lamp 33 in a direction parallel to the platen glass 31. Scans the original on the platen glass 31. The reflected light from the original is reflected by a mirror 34, and then further reflected by mirrors 35 and 36.
CCD image sensor 3 through a condenser lens 37
8 and is converted into an electric signal to generate image data.

The image data is A / D-converted into a digital signal by the control unit 100, and after being subjected to a known correction process such as shading correction, is subjected to a rotation process in accordance with the result of a direction determination process described later. , A drive signal for the laser diode 51 of the printer unit 50. The printer unit 50 forms an image on paper by a known electrophotographic method, and upon receiving the drive signal, drives the laser diode 51 to emit laser light.
The laser light is applied to a polygon mirror 5 rotating at a predetermined angular velocity.
The light is reflected by the two mirror surfaces, and the surface of the photosensitive drum 56 is exposed and scanned through the fθ lens 53 and the mirrors 54 and 55.

Before the photosensitive drum 56 receives the above-mentioned exposure, the cleaning unit 57 removes residual toner on the surface of the photosensitive drum, and after being irradiated with an eraser lamp (not shown) to be neutralized, the charged drum is charged. The photosensitive drum 56 is charged uniformly, and when the above-described exposure is performed in such a state of being uniformly charged, an electrostatic latent image is formed on the surface of the photosensitive drum 56.

The developing unit 59 develops the electrostatic latent image formed on the surface of the photosensitive drum 56. On the other hand, the paper supply unit 70 is provided with two paper cassettes 71 and 72,
The paper is fed from one of the paper cassettes 71 and 72 by driving the paper feed roller 711 or 721 in synchronization with the exposure and development operations on the photosensitive drum 56 described above. The fed sheet comes into contact with the surface of the photosensitive drum 56 below the photosensitive drum 56, and at this time, the toner image formed on the surface of the photosensitive drum 56 is moved by the electrostatic force of the transfer charger 60. It is transferred to the paper surface.

Thereafter, the sheet is separated from the surface of the photosensitive drum 56 by the electrostatic force of the separation charger 61, and is conveyed to the fixing unit 63 by the conveying belt 62. The toner image transferred to the paper is fixed by being pressed while being heated by a fixing roller 64 having a heater inside in a fixing unit 63. After fixing, the paper is fed to a discharge roller 65.
Is discharged onto the discharge tray 66.

An operation panel 90 is provided at an easy-to-operate position on the front surface of the image reading section 30. A numeric keypad for inputting the number of copies, a start key for instructing the start of copying, and an original for easy filing. The setting key for setting various copy modes such as a binding margin mode in which the image is shifted in a predetermined direction by a predetermined amount (shifted) and copied on paper to set a binding margin mode, and the above setting keys. A display unit or the like for displaying the set mode by a message is provided.

(2) Configuration of the control unit 100 FIG. 2 is a block diagram showing the configuration of the control unit 100 installed inside the copying machine. As shown in the figure, the control unit 100 includes an image signal processing unit 10
2. Compression unit 103, image memory 104, image rotation unit 10
5. Decompression unit 106, LD drive unit 107, top and bottom recognition unit 10
9. Top / bottom recognition result memory 110, ROM 111, RAM
112 and the like.

The image signal processing unit 102 includes the image reading unit 30
The image data sent from the CCD sensor 38 is subjected to various types of correction processing such as known shading correction and MTF correction, and is sent to the compression unit 103 and the top and bottom recognition unit 109. The compression unit 103 compresses the image data sent from the image signal processing unit 102 in page units and stores the compressed data in the image memory 104.

Upon receiving a read request from the CPU 101, the image rotation unit 105 reads image data of a target page from the image memory 104, rotates the image data as necessary according to a rotation angle signal from the CPU 101, The data is sent to the extension unit 106. The decompression unit 106 performs decompression processing on the image data sent from the image rotation unit 105 and sends the image data to the LD driving unit 107. In the present embodiment, the fixed-length compression method is used as the compression method, and the decompression processing is performed after performing the rotation processing on the compressed image data. However, the present invention is not limited to this. The rotation processing may be performed after the data is expanded.

The LD driving section 107 generates a driving signal for driving the laser diode (LD) 51 based on the expanded image data. The top / bottom recognition unit 109 recognizes the orientation of the image (that is, the top / bottom direction) of each document based on the image data sent from the image signal processing unit 102, and calculates the reliability of the orientation.

More specifically, first, one page of the document is divided into a plurality of areas, and a known top-and-bottom recognition process is performed on each area. There are various methods for this upside-down recognition processing (Japanese Patent Application Laid-Open Nos. 4-22976 and 7-6512).
No. 0, etc.) has been published, so a detailed description will be omitted, but the basic procedure is as follows. First, data of one character is cut out from image data for one area divided as a processing target area in accordance with a histogram, and character data (characters for comparison) corresponding to the cut out data is obtained.
Is found from a pattern dictionary in a memory outside the figure. Then, the character for comparison is rotated by 90 degrees and compared with the cut-out data. Then, the angle (0, 90, 180 or 270 degrees) at the time of matching is obtained as information indicating the direction of the cut-out character, and the direction of the image in the area is determined based on the information.

The top / bottom recognition unit 109 sequentially determines the orientation of the image with respect to each area, checks the number of areas having the same orientation, and determines the orientation of the image in the area where the number is largest to the image of the original. (Hereinafter, referred to as “image direction”). For example, when one page of a document is divided into nine regions (P1 to P9) substantially equally, the direction of the cut-out character is 0 degree for the regions P1 to P8 and 90 degrees for the region P9. For example, the orientation of the image of the document is recognized as 0 degree.

Then, the degree of reliability P with respect to the direction of the image of the recognized document is calculated. Here, the reliability P (%) =
(Maximum number of areas where character orientations match / Number of all areas) x 10
In the above example, the reliability P is 89%. If the document contains only characters in a line, it is easier to extract characters from the histogram, which increases the reliability.However, if the document contains information other than characters such as ruled lines and figures, characters are difficult to extract from the histogram, making it difficult to extract characters from the histogram. In some cases, the direction of the document cannot be determined.

As another method for obtaining the orientation of an image and its reliability, a technique described in JP-A-2000-32247 can be used. Top and bottom recognition unit 10
9 receives the instruction of the top and bottom recognition from the CPU 101,
Based on the image data of the document sent from the image signal processing unit 102, the orientation of the image is recognized and the reliability P for the recognition result is calculated, and the number of pages of the document and the direction of the image (specifically, Is stored in the management table of the top / bottom recognition result memory 110 in association with the angle data such as 0 degree or 90 degree) and the value of the reliability P.

The ROM 111 stores a program relating to a scanning operation in the image reading unit 30 and an image forming operation in the printer unit 50, and a program for a direction determination process to be described later. The RAM 112 temporarily stores various control variables, a print flag described later, and the value of the variable n.

The CPU 101 controls the operation of each unit such as the automatic document feeder 10, the image reading unit 30, the printer unit 50, etc. based on the control program of the ROM 111, and realizes a smooth image forming operation. Further, a direction determination process is performed. In this process, when the reliability calculated by the top / bottom recognition unit 109 with respect to image data of a certain document exceeds a predetermined value, the orientation of the image recognized with respect to the document is uniformly changed for all the documents in the document bundle. This is determined by considering the orientation of the image. When the orientation of the image is determined, the value of the print flag is changed from "0" to "1" to start the image forming operation.

When the direction of the image is determined by the direction determination processing, the CPU 101 stores data relating to the direction, specifically, data indicating an angle such as 0 ° or 90 ° in the RAM 1.
12 is stored. (3) Control Operation of Control Unit 100 FIG. 3 is a flowchart showing a main routine of the control operation of control unit 100.

When the power of the copier is turned on, first, the CPU
Reference numeral 101 denotes an initial setting for clearing the contents of the RAM 112 and the top / bottom recognition result memory 110 in the control unit 100, setting the value of the print flag to "0", and setting the value of the variable n to "1", and performing copy. A mode initialization process is performed (step S1). Subsequently, the internal timer is started in step S2. An internal timer sets the processing time of one of the main routines.

When the user presses the start key on the operation panel 90 ("Y" in step S3), the CPU 101 sets the document feed tray 11
In addition to carrying out transport control for sequentially transporting a plurality of (M) documents placed on the document to the image reading position one by one, reading control for reading an image of the transported document is performed (step S5). In the present embodiment, it is assumed that pages from page 1 to page M are added to each document, and the reading operation is performed sequentially from the first page.

The read image data of the document is temporarily stored in the image memory 104 of the control section 100 temporarily.
Further, the CPU 101 instructs the top and bottom recognition unit 109 until the orientation of the image is determined by the
The orientation of the image is recognized in order from the document on the page, the reliability P for the result is calculated, and the data is stored in the top-bottom recognition result memory 110.

Next, the CPU 101 performs a direction determination process (step S5). Details of this processing will be described later. Subsequently, the CPU 101 performs an image forming process (step S6). First, the value of the print flag is referred to, and if this is "1", the image forming operation is started. That is, the sheet feeding unit 70 is controlled to start a sheet feeding operation necessary for image formation, and the image rotating unit 105 is controlled to sequentially transfer image data from the image data of the first page document to the image memory. 104. And RAM
The content of the determined orientation of the image stored in 112 is checked, and if it is necessary to rotate the image data to form an image according to the orientation of the image, the rotation angle is determined by a rotation angle signal. To rotate the image data.

For example, when the binding margin mode is set by the user, the determined image direction (top and bottom direction) and the paper feeding unit are set so that a margin portion as a binding margin is formed at an appropriate position. The angle to be rotated is calculated according to the direction (vertical or horizontal) of the sheet fed from 70, and the image rotation unit 105 is instructed to rotate the image data at the angle.

After the rotated image data is expanded by the expansion unit 106, it is sent to the LD driving unit 107, where the laser diode 51 is driven. The CPU 101 controls the printer unit 50 in parallel with this operation to perform an image forming operation. If the print flag is “0”, the image forming operation cannot be started because the direction of the image has not been determined in the direction determination processing as described later, and the process proceeds to the next step S7.

In step S7, it is determined whether or not the image forming operation for all the originals has been completed. If it is determined that the image forming operation has not been completed ("N" in step S7), the process proceeds to step S8. In step S8, the process returns to step S2 after waiting for the end of the internal timer. In step S3, the process proceeds to step S4 until the end of the image forming operation is determined in step S7, whereby the routine circulates during the copying operation.

On the other hand, if it is determined in step S 7 that the image forming operation has been completed, the data of the determined image orientation stored in the RAM 112 and each document in the management table of the top / bottom recognition unit 109 are compared. Reset all the data such as the data of the recognized image orientation (step S9), set the print flag to "0" (step S10), and proceed to step S8.

FIG. 4 is a flowchart showing a subroutine of the direction determining process in step S5. As shown in the figure, the CPU 101 first determines whether or not the image orientation has been determined (step S11). This determination can be made by referring to the value of the print flag. That is,
If the orientation of the image has been determined, the print flag is "1"
If the print flag is "1", the orientation of the image has already been determined, and the process returns to the main routine. If there is, the process moves to step S12. For the sake of explanation, here, it is assumed that the original of the first page has just been read and the orientation of the image has not yet been determined, and the process proceeds to step S12.

In step S12, it is determined whether or not the recognition of the orientation of the image has been completed for the document of page n (1 ≦ n ≦ M). First, it is determined whether or not the recognition process of the orientation of the document image by the top and bottom recognition unit 109 has been completed for the document of the first page (n = 1). This determination is made based on whether or not the recognition result is stored in the management table of the top and bottom recognition result memory 110. If the recognition result has not been stored yet, the recognition processing has not been completed, that is, the recognition processing is being performed. ("N" in step S12), and returns to the main routine.

On the other hand, if it is determined that the recognition processing has been completed, that is, that the recognition result is stored (“Y” in step S12), the orientation of the image recognized with respect to the n-page document is determined. It is determined whether or not the reliability Pn is larger than a predetermined value K. Here, the predetermined value K is used as a threshold value for determining whether or not the orientation of the recognized image is appropriate, and is stored in the ROM 111 in advance. In the present embodiment, K is set to 80 (%). The reliability Pn for the n-page document can be obtained by referring to the management table in the top / bottom recognition result memory 110.

If the value of the reliability P1 for the original of the first page (n = 1) exceeds the value of K ("Y" in step S13), the original image of the first page is recognized. The determined orientation is determined as the orientation of the image of all originals (all originals from page 1 to page M) (step S14). Then, the value of n is reset to "1" (step S15), the print flag is set to "1" (step S16), and the process returns to the main routine.
In addition, it instructs the top / bottom recognition unit 109 to stop the process of recognizing the orientation of the image of the document of the second and subsequent pages and calculating the reliability.

In step S16, the print flag is "1".
, The image forming operation is started in the next step S6. On the other hand, if P1 ≦ K (“N” in step S13), it is determined whether there is a document of the next page (step S17). Here, since the second page document exists, the process proceeds to step S18. In step S18, after the current value of n is incremented by "1", the process returns to the main routine. Thus, the value of n becomes “2”.

In the series of processing of "N" in step S13, "Y" in S17, and S18, the value of the print flag remains "0" because the orientation of the image was not determined. When determining that the print flag is “0” in step S6, the CPU 101 does not start the image forming operation, and proceeds directly to step S7. The routine makes one round, and the process again proceeds to the direction determination determination process of step S5. If “Y” is obtained in step S12, step S5
In step 13, the value of the reliability P2 for the document of the second page (n = 2 in step S18) is compared with a predetermined value K. Here, if P2> K, the process proceeds to step S14, where the orientations of the images of all the originals are determined to be the directions recognized with respect to the image of the original on the second page, and the value of n is set to “1”. (Step S1)
5) After setting the print flag to “1” (step S1)
6) Return to the main routine. Thus, in the next step S6, the image forming operation is sequentially started from the original of the first page. In addition, it instructs the top / bottom recognition unit 109 to stop the recognition of the orientation of the image of the document on the third and subsequent pages and the calculation processing of the reliability.

On the other hand, for the original of the second page, P2
If .ltoreq.K ("N" in step S13), the processes in S17 and S18 are performed to perform the same process on the third page of the document as in the second page, and then the process returns to the main routine. Of course, also in this case, the image forming operation is not started. This process is repeatedly performed up to the last M-page document. When Pn> K, the direction recognized for the image of the n-th document is determined as the orientation of the image of all the documents. Then, in step S6, the image forming operation is sequentially started from the original of the first page. In addition, instruct the top and bottom recognition unit 109,
The process of recognizing the orientation of the image of the document after the (n + 1) th page and calculating the reliability is stopped.

If Pm> K is not satisfied even if the above processing is performed up to the final M page document (“N” in step S13, “N” in S17), For example, 0 ° is determined as the orientation of the images of all the originals (step S19), and the process proceeds to step S15. As described above, in the present embodiment, as described above, it is generally assumed that each document of the document bundle set by the user in the automatic document feeder 10 is in the same direction in many cases. The top and bottom recognition processing is performed on the originals in the reading order, and when Pn> K for the n-th original, the orientation of the image is determined to be the orientation of the images of all the originals. .

Thus, for example, if the first page is a cover page containing a chart and a business document or the like in which a table of contents or a text having only character strings is read in the manuscript of the second page or later, there is a chart. For the cover, since the reliability is generally low, the relationship of P> K does not hold, so that the orientation recognized for the original of the first page is not adopted and the reliability increases. The orientation of the image is determined for the document on the second page.

Therefore, in a case where the business document is copied using the binding margin mode, even if the user sets the document bundle in the wrong direction, the top and bottom of the image of the document on the page with high reliability is properly adjusted. Because it is judged, simply 1
As compared with the third related art in which the top and bottom recognition is performed only for the document on the page, errors in the top and bottom recognition can be greatly reduced. When the reliability P exceeds the predetermined value K, the image forming operation is started, and the top and bottom recognition processing is not performed on the document of the subsequent pages. Therefore, the image data of all the documents is temporarily stored in the memory. And performs the top / bottom recognition process on each document. The second prior art performs the top / bottom recognition process on each document and performs an image forming operation on the document each time the top / bottom is recognized. As compared with the prior art, the start of the image forming operation can be accelerated, and the time required for the image forming operation can be shortened. As a result, a decrease in copy productivity can be suppressed.

(4) Modifications Although the present invention has been described based on the embodiments,
The present invention is, of course, not limited to the above-described embodiment, and the following modifications are conceivable. (4-1) In the above embodiment, if the reliability P does not exceed the predetermined value K up to the last page of the document, the direction of the image is determined to be the predetermined direction (step S1).
7) The method is not limited to this method. For example, by referring to the management table of the top / bottom recognition result memory 110, the number of documents having the same orientation is fixed at a fixed rate,
%, The orientation of the image may be determined as the orientation of the images of all the originals.

If it is important to accelerate the start of the image formation, even if the reliability P does not exceed the predetermined value K, for example, when the documents having the same orientation are continuous (the first page and the second page). If they are the same on the page)
Alternatively, the direction can be determined. (4-2) In the above-described embodiment, 1 is read first.
Although the process of recognizing the orientation of the image is performed from the document of the page, the present invention is not limited to this method. For example, the process may be performed from the document of the second page. In the above-mentioned business documents and the like, it is general that the reliability of the cover of the first page is low. Therefore, if the recognition process is not performed for the first page and the process is performed from the second page, the reliability is low. The time required for the recognition processing of the first page, which is expected to be, can be saved, and the processing time can be shortened.

Further, it is not necessary to limit the recognition process to the reading order of the originals (the page order in the above description). For example,
Even if the recognition process is performed on odd pages such as 1, 3, 5,..., The effect of reducing the recognition processing time can be obtained as compared with the conventional case where the recognition process is performed on all originals. (4-3) In the above embodiment, an example in which the image recognition device according to the present invention is applied to a digital copying machine has been described. However, the present invention is applicable to other devices that need to read an original, such as a facsimile device and an image It is also applied as an image recognition device in a reader. Further, the direction of the image may be recognized based on the image data of the document received from an image input device such as another image reader. In this case,
The number of transmitted documents may be determined by receiving a signal indicating the end of transmission of image data from the image input device or by receiving information indicating the number of transmitted documents.

[0053]

As described above, according to the image recognition apparatus of the present invention, the orientation of the image of a document recognized for a document whose reliability exceeds a predetermined value is determined as the image orientation of all the documents. Therefore, errors in image recognition can be reduced as compared with the conventional case where the orientation of the image is recognized only for the original document that has been read first. When the reliability exceeds a predetermined value, the image orientation recognition process is not performed for the subsequent documents, so that it is not necessary to perform the image orientation recognition for all the documents, and the time required for recognition can be reduced. Therefore, if the image recognition device is applied to an image forming apparatus, the start of the image forming operation can be accelerated and the time required for the image forming operation can be shortened as compared with the related art.

Since the recognition of the orientation of the image is started from the image data of one page of the original document received first, the recognition of the orientation of the image is started more than the recognition of the image direction from the image data of the original document with the lower reception order. The direction can be determined quickly. Furthermore, until the reliability becomes larger than the predetermined value, the image data of each page is recognized in the order of reception of the image data, so that the recognition of the image direction is not performed in the order of reception. There is an effect that the direction of the image can be determined in the shortest time.

[Brief description of the drawings]

FIG. 1 is a diagram showing an entire configuration of a copying machine according to the present invention.

FIG. 2 is a block diagram showing a configuration of a control unit installed inside the copying machine.

FIG. 3 is a flowchart showing a main routine of a control operation of the control unit.

FIG. 4 is a flowchart illustrating a subroutine of a direction determination process.

[Explanation of symbols]

 REFERENCE SIGNS LIST 10 automatic document feeder 30 image reading unit 50 printer unit 70 paper feed unit 90 operation panel 100 control unit 101 CPU 102 image signal processing unit 103 compression unit 104 image memory 105 image rotation unit 106 expansion unit 107 LD drive unit 109 top and bottom recognition unit 110 Top and bottom recognition result memory 111 ROM

Continuation of the front page (72) Inventor Kazuyuki Kawabata 2-3-13 Azuchicho, Chuo-ku, Osaka City, Osaka Inside Osaka International Building Minolta Co., Ltd. (72) Inventor Kazuhiro Ueda 2-chome Azuchicho, Chuo-ku, Osaka City, Osaka No.3-13 Osaka International Building Minolta Co., Ltd. F term (reference) 2C087 AA09 AC08 BA03 BD01 BD22 BD40 5B057 CA18 CB16 CD02 CD04 CH11 DA20 DB02 DC08 5C076 AA17 AA24 5L096 BA07 BA17 CA17 FA14 FA44 FA67 GA19 GA30 HA09 JA24

Claims (7)

[Claims] 1. An image recognition device that receives image data of a plurality of originals and recognizes the orientation of an image, comprising: a recognition unit that recognizes an orientation of an image of the original based on image data of a predetermined page. Calculating means for calculating the reliability of the recognition result of the recognition means; if the calculated reliability is equal to or less than a predetermined value, another one page until the reliability becomes larger than the predetermined value. Control means for controlling the recognizing means so as to recognize the orientation of the image of the document based on the image data of the document; and if the calculated reliability exceeds the predetermined value, recognition is performed on the document. Determining means for determining the image orientations of all the received documents as image orientations. 2. The image data for one predetermined page,
2. The image recognition apparatus according to claim 1, wherein the image data is one page of image data of a document received first. 3. The image processing apparatus according to claim 1, wherein the control unit causes the image data of one page to be recognized in the order of reception of the image data until the reliability becomes larger than the predetermined value. Or the image recognition device according to 2. 4. The determination means according to claim 1,
4. The image recognition apparatus according to claim 1, wherein when the reliability is equal to or less than the predetermined value, a predetermined direction is determined as the directions of the images of all the originals. . 5. The method according to claim 1, wherein the determining unit refers to a direction of the image recognized for each document, and determines that the document recognized as having the same orientation is continuously received for a predetermined number of pages. 5. The image recognition apparatus according to claim 1, wherein even if the reliability does not exceed the predetermined value, the orientation is determined as the orientation of images of all the originals. 6. An image recognition apparatus according to claim 1, further comprising: an image data rotation unit configured to perform a rotation process on the received image data based on a determination result of the determination unit of the image recognition apparatus. An image forming apparatus comprising: an image forming unit that forms an image based on image data that has been subjected to a rotation process. 7. The image forming device according to claim 6, wherein the image forming device starts the image forming operation when the reliability calculated by the calculating device of the image recognition device exceeds a predetermined value.
An image forming apparatus according to claim 1.