Image reading device

Abstract

To allow a determination of a defect in a document image highly convenient for a user.SOLUTION: An image forming apparatus 10 comprises an image reading unit 11, a control unit 31, a sheet defect determination unit 32, a description information defect determination unit 33, and a determination unit 34. The image reading unit 11 reads an image of a document. The control unit 31 causes the image reading unit 11 to sequentially read the document one by one. The sheet defect determination unit 32 determines the presence or absence of a defect in a sheet shape of the read document on the basis of a document image obtained through document reading performed by the image reading unit 11. The description information defect determination unit 33 determines the presence or absence of a defect in description information of the read document on the basis of the document image obtained through reading performed by the image reading unit. When the sheet defect determination unit 32 determines the presence of a defect in the sheet shape of the document, and when the description information defect determination unit 33 determines the presence of a defect in the description information of the document, the determination unit 34 determines the presence of a defect in the document.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to an image reading apparatus for reading an image of a document.

Background Art

[0002] In an image forming apparatus as an example of an image reading apparatus, an image of a document is read and the image of the document is formed on a recording sheet. However, when the document has a corner fold, the corner of the read image of the document is chipped, and the corner of the image of the document formed on the recording sheet is also chipped.

[0003] The image reading apparatus described in Patent Document 1 includes a light receiving unit that receives reflected light from a document, and a first light emitting unit and a second light emitting unit that are respectively arranged on the upstream side and the downstream side with the light receiving unit interposed therebetween in the reading direction of the document. The first light emitting unit reads an image of the document (first image) in a state where the first light emitting unit is lit, the second light emitting unit reads an image of the document (second image) in a state where the second light emitting unit is lit, and the first light emitting unit and the second light emitting unit read an image of the document (third image) in a state where both are lit. The image processing unit compares the first image and the second image to determine a fold line in the center of the document, and corrects the density values of the pixels in the range affected by the fold line based on the third image.

[0004] Further, in the image reading apparatus described in Patent Document 2, a sticker information extraction unit that extracts sticker information regarding a sticker from a document image with a sticker, and an image processing unit that separately generates a stickerless document image obtained by excluding the sticker from the document image with a sticker and a sticker information image in which the sticker information regarding the stickerless document image is described based on the sticker information. The document with the sticker is output as a stickerless document image, and the sticker information image is also separately output so that it can be confirmed.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0006] When the manuscript has a corner fold as described above, the image of the read manuscript is also missing. However, even if the manuscript has a corner fold, if the influence of the corner fold does not affect the description information (such as text) of the manuscript, the image of the manuscript can be used without problems. Considering the trouble of correcting the corner fold and re-reading the image of the manuscript, it may be better not to correct the corner fold and re-read the image of the manuscript. Therefore, even if the manuscript has a corner fold, if there is no loss in the description information of the manuscript, it is desirable to be able to use the image of the read manuscript.

[0007] In Patent Document 1, the fold in the center of the manuscript is determined, and the density values of the pixels in the range affected by the fold are corrected. However, as described above, although the manuscript has a corner fold, the processing in the case where there is no loss in the description information of the manuscript is not disclosed.

[0008] Also, in Patent Document 2, a manuscript with a sticker is output as a manuscript image without a sticker. However, again, the processing in the case where the manuscript has a corner fold but there is no loss in the description information of the manuscript is not disclosed.

[0009] Therefore, the present invention has been made in view of the above circumstances, and an object thereof is to enable a user to easily determine the loss of a manuscript image.

Means for Solving the Problems

[0010] An image reading apparatus according to an aspect of the present invention includes an image reading unit that reads an image of a document, a control unit that causes the image reading unit to sequentially read documents one by one, a paper defect determination unit that determines whether there is a defect in the paper shape of the read document based on the document image obtained by reading the document by the image reading unit, a description information defect determination unit that determines whether there is a defect in the description information of the read document based on the document image obtained by reading by the image reading unit, and a determination unit that determines that there is a defect in the document when the paper defect determination unit determines that there is a defect in the paper shape of the document and the description information defect determination unit determines that there is a defect in the description information of the document.

Effect of the Invention

[0011] According to the present invention, even when there is a defect in the paper shape of the document, if there is no defect in the description information of the document, by making the image of the read document an effective image, it becomes possible to determine the defect of the document image that is highly convenient for the user.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3A

Figure 3B

Figure 3C

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Embodiments for Carrying Out the Invention

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a cross-sectional view showing an image forming apparatus 10 which is an embodiment of an image reading apparatus according to the present invention. As shown in FIG. 1, the image forming apparatus 10 of the present embodiment is an MFP (multifunction peripheral) having a plurality of functions such as a copying function and a scanner function. This image forming apparatus 10 includes an image reading unit 11 and an image forming unit 12.

[0014] The image reading unit 11 has an image pickup device (CCD sensor, CIS) that optically reads an image of the original manuscript M, and an original image of the original manuscript M is generated from the analog output of this image pickup device. Also, in the image reading unit 11, the image of the original manuscript M placed on the platen glass is read by the image pickup device, or a plurality of original manuscripts M set in the original tray by the ADF (Automatic Document Feeder) are pulled out one by one and conveyed, and the images of the conveyed original manuscripts M are sequentially read by the image pickup device.

[0015] The image forming unit 12 forms the image of the document M indicated by the above image data on the recording paper, and includes an image forming unit 3M for magenta, an image forming unit 3C for cyan, an image forming unit 3Y for yellow, and an image forming unit 3Bk for black. In any of the image forming units 3M, 3C, 3Y, and 3Bk, the surface of the photosensitive drum 4 is uniformly charged, the surface of the photosensitive drum 4 is exposed to form an electrostatic latent image on the surface of the photosensitive drum 4, the electrostatic latent image on the surface of the photosensitive drum 4 is developed into a toner image, and the toner image on the surface of the photosensitive drum 4 is primarily transferred to the intermediate transfer belt 5. As a result, a color toner image is formed on the intermediate transfer belt 5. This color toner image is secondarily transferred to the recording paper P conveyed through the conveyance path 8 from the paper feeding unit 14 in the nip area N between the intermediate transfer belt 5 and the secondary transfer roller 6.

[0016] Thereafter, the recording paper P is heated and pressurized by the fixing device 15, the toner image on the recording paper P is fixed by thermocompression bonding, and the recording paper P is further discharged to the discharge tray 17 through the discharge roller 16.

[0017] FIG. 2 is a block diagram showing the main internal configuration of the image forming apparatus 10 of the present embodiment. As shown in FIG. 2, the image forming apparatus 10 of the present embodiment includes an image reading unit 11, an image forming unit 12, a display unit 21, an operation unit 22, a touch panel 23, an image memory 26, a storage unit 28, and a control unit 29. These components are capable of transmitting and receiving data or signals to and from each other through a bus.

[0018] The display unit 21 is composed of a liquid crystal display (LCD) or an organic EL (OLED) display. The operation unit 22 includes physical keys such as a numeric keypad, an enter key, and a start key.

[0019] On the screen of the display unit 21, a touch panel 23 is arranged. The touch panel 23 is a touch panel such as a so-called resistive film method or a capacitance method, and detects contact (touch) of a user's finger or the like with the touch panel 23 together with the contact position, and outputs a detection signal indicating the coordinates of the contact position to a control unit 31 (to be described later) of the control unit 29 and the like.

[0020] In the image memory 26, the original image of the original document read by the image reading unit 11 is stored.

[0021] The storage unit 28 is a large-capacity storage device such as an SSD (Solid State Drive) or an HDD (Hard Disk Drive), and stores various application programs and various data.

[0022] The control unit 29 is composed of a processor, a RAM (Random Access Memory), a ROM (Read Only Memory), and the like. The processor is, for example, a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or an MPU (Micro Processing Unit), etc. The control unit 29 functions as a control unit 31, a paper shortage determination unit 32, a description information shortage determination unit 33, and a determination unit 34 when the control program stored in the above ROM or the storage unit 28 is executed by the above processor.

[0023] The control unit 31 comprehensively controls the image forming apparatus 10. The control unit 29 is connected to the image reading unit 11, the image forming unit 12, the display unit 21, the operation unit 22, the touch panel 23, the image memory 26, the storage unit 28, and the like. The control unit 31 performs operation control of these components and transmission and reception of signals or data between each of these components.

[0024] The control unit 31 serves as a processing unit that executes various processes necessary for image formation by the image forming apparatus 10. Further, the control unit 31 receives an operation instruction input by the user based on a detection signal output from the touch panel 23 or an operation of a physical key of the operation unit 22. For example, the control unit 31 receives a touch operation on a GUI (Graphical User Interface) or the like displayed on the screen of the display unit 21 through the touch panel 23. Furthermore, the control unit 31 has a function of controlling the display operation of the display unit 21.

[0025] Also, the paper shortage determination unit 32 determines the presence or absence of a shortage in the paper shape of the original document M based on the original document image obtained by reading the original document M by the image reading unit 11. The description information shortage determination unit 33 determines the presence or absence of a shortage in the description information of the original document M based on the image data. The determination unit 34 determines that there is a shortage in the original document M when the paper shortage determination unit 32 determines that there is a shortage in the paper shape of the original document M and the description information shortage determination unit 33 determines that there is a shortage in the description information of the original document M.

[0026] In the image forming apparatus 10 having such a configuration, for example, the user sets a plurality of original documents M in the image reading unit 11 and operates the start key of the operation unit 22. As a result, an original document reading instruction is received by the operation unit 22. The control unit 31 causes the image reading unit 11 to sequentially read the images of the respective original documents M according to the original document reading instruction, and sequentially stores the original document images of the respective original documents M in the image memory 26. Further, the control unit 31 causes the image forming unit 12 to form the original document images of the respective original documents M on the respective recording papers.

[0027] Further, the control unit 31 operates the image forming apparatus 10 in a defect detection mode in which a defect of the document M is detected each time an image of each document M is sequentially read by the image reading unit 11. In the case of the operation in the defect detection mode, each time an image of each document M is sequentially read by the image reading unit 11, the paper defect determination unit 32 determines whether there is a defect in the paper shape of the document M, and the described information defect determination unit 33 determines whether there is a defect in the described information of the document M. Then, when it is determined that there is a defect in the paper shape of the document M and it is determined that there is a defect in the described information of the document M, the determination unit 34 determines that there is a defect in the document M. At this time, the control unit 31 interrupts the reading of the document M by the image reading unit 11 at the time of the determination by the determination unit 34 and causes a warning to be displayed on the display unit 21, or does not stop the reading of the document M by the image reading unit 11 at the time of the determination by the determination unit 34, and causes a warning to be displayed on the display unit 21 after all the documents M have been read by the image reading unit 11.

[0028] Next, the process of determining a defect of the document M based on the document image of each document M by the image reading unit 11 as described above will be described with reference to the flowcharts shown in FIGS. 3A, 3B, and 3C.

[0029] The user operates a GUI (not shown) displayed on the display unit 21 to input a setting instruction for setting the defect detection mode. The operation unit 22 receives the setting instruction. When the control unit 31 detects an instruction for setting the defect detection mode through the touch panel 23, it sets the operation mode of the image forming apparatus 10 to the defect detection mode (S101). The control unit 31 causes a setting screen G1 for the defect detection mode as shown in FIG. 4 to be displayed on the display unit 21 (S102).

[0030] On the setting screen G1 for the defect detection mode, there is a key K11 for inputting a defect-time interruption instruction for selecting to interrupt the document reading by the image reading unit 11 at that time when a defect of the document M read by the image reading unit 11 is detected, and even if a defect of the document M is detected, the document reading by the image reading unit 11 is not interrupted at that time, the operation of the image reading unit 11 is continued, and a key K12 for inputting a defect-time continuation instruction for reading images of all the documents M set in the image reading unit 11 is displayed.

[0031] Here, it is assumed that the user touches one of the keys K11 and K12 on the setting screen G1 in the defect detection mode, sets a plurality of originals M in the image reading unit 11, and operates the start key of the operation unit 22.

[0032] When the operation unit 22 receives an interruption instruction at the time of defect due to the operation of the key K11 by the above user, and receives an execution instruction for the copy operation in the operation unit 22 by the operation of the start key (S103 "interruption"), the control unit 31 stores the interruption instruction at the time of defect (S104). Also, when the operation unit 22 receives a continuation instruction at the time of defect from the operation of the key K12 by the user, and receives an execution instruction for the copy operation in the operation unit 22 by the operation of the start key (S103 "continuation"), the control unit 31 stores the continuation instruction at the time of defect (S105).

[0033] Then, the control unit 31 starts reading the original M by the image reading unit 11 one by one in response to the start key of the operation unit 22 (S106). When a document image is obtained by reading the original M by the image reading unit 11, the document image of the original M is stored in the image memory 26 (S107).

[0034] Each time a document is read one by one, the paper defect determination unit 32 of the control unit 31 determines whether there is a defect in the paper shape of the document M based on the document image of each document M stored in the image memory 26 (S108). For example, as a method for detecting a defect in the paper shape, a method based on deep learning (deep neural network) using a CNN (Convolution Neural Network) is applied. A CNN is a type of multi-layer neural network that includes a layer specialized for image processing within the network. Various networks have been devised for CNN depending on its purpose. For example, there are networks that perform image classification to classify an image into multiple classes by identifying the type of an object shown in the image or the image, and semantic segmentation to extract a specific area from an image on a pixel-by-pixel basis. Here, the task of "detecting a defect in the paper shape from the image of the read document M" is regarded as the task of "classifying the image of the read document M into two types: with and without a defect in the paper shape", and a model (algorithm) is created that detects and outputs a defect in the paper shape by deep learning using an image classification network. Deep learning of CNN is performed by giving a data group called training data to the network. The training data is composed of a pair of input image data and classification result data (correct answer data).

[0035] In this embodiment, the input image includes binary data of a document image (0) of a document M with no missing paper shape as shown in, for example, FIG. 5(A), and a document image (1) of a document M with a missing paper shape due to a corner fold as shown in FIGS. 5(B) and 5(C). By preparing a large amount of such training data and training the network, a classification model is generated to classify (determine) whether there is a missing paper shape in the given image of document M. Then, by using the classification model, the missing paper shape of the read document M, that is, the missing paper shape due to the corner fold of the paper of document M is determined. Also, by training the network with a document image of a document M with no part of the paper missing as shown in FIG. 6(A), and a document image of a document M with a part of the paper torn and missing as shown in FIGS. 6(B) and 6(C), it may be determined that the document M has a missing part where the paper is torn and missing.

[0036] Alternatively, the paper defect determination unit 32 of the control unit 31 detects the contour of document M based on the document image of document M stored in the image memory 26, extracts four sides orthogonal to each other forming the detected contour, extracts the inclined sides inclined with respect to the four sides at the corners of the document image if there are any, and when the inclined sides are equal to or greater than a preset threshold value, it may detect that there is a missing paper shape of document M, that is, the corner of the paper of document M is folded.

[0037] Note that any one of the above detection methods by the paper defect determination unit may be selectable by a user's GUI operation. For example, the control unit 31 displays respective keys associated with each detection method on a setting screen G1 of a defect detection mode shown in FIG. 4, detects a touch operation on any one of the keys through the touch panel 23, and sets the detection method associated with the detected key.

[0038] When the paper defect determination unit 32 determines that there is a defect in the paper shape of the document M (S108 "Yes"), the described information defect determination unit 33 determines whether there is a defect in the described information of the document M based on the document image of the document M stored in the image memory 26 (S109). That is, each time a document is read one by one and the result in S108 is Yes, the described information defect determination unit 33 determines whether there is a defect in the described information of the document M. For example, as a method for detecting a defect in the described information of the document M, a method based on deep learning using a CNN is also applied. In this case, the problem of "detecting a defect in the described information from the image of the read document" is regarded as the problem of "classifying the image of the read document M into two types: with and without a defect in the described information", and a model (algorithm) capable of detecting a defect in the described information is created and applied by deep learning using an image classification network. In the present embodiment, the input image includes binary data such as an image (0) of a document M without a defect in the described information such as text as shown in FIGS. 7(A) and 7(B), and an image (1) of a document M with a defect in the described information such as text due to a corner fold as shown in FIGS. 7(C) and 7(D). By preparing a large amount of such training data and training the network, a classification model for classifying (determining) whether there is a defect in the described information in the given image of the document M is generated. Then, by using the classification model, a defect in the described information of the read document M, that is, a defect in the described information of the document M due to a corner fold is determined. Further, by training the network with an image of a document M without a sticker attached as shown in FIG. 8(A) and images of a document M with a sticker H attached as shown in FIGS. 8(B) and 8(C), a defect in the described information of the document M due to the sticker H may be detected.

[0039] Note that any one of the above-described detection methods by the described information defect determination unit 33 may be made selectable by a user's GUI operation. For example, the control unit 31 displays each key associated with each detection method on the setting screen G1 of the defect detection mode shown in FIG. 4, detects a touch operation on any key through the touch panel 23, and sets the detection method associated with the detected key.

[0040] When the description information deficiency determination unit 33 determines that there is no deficiency in the description information of the manuscript M (S109 "No"), the control unit 31 determines whether a next manuscript M that has not yet been read is detected by a sensor (not shown) of the image reading unit 11 (S110). That is, the control unit 31 determines whether there is a next manuscript M that has not yet been read. Here, when the control unit 31 determines that there is a next manuscript M (S110 "Yes"), S107 to S110 are repeated.

[0041] Also, when the paper deficiency determination unit 32 determines that there is no deficiency in the paper shape of the manuscript M (S108 "No"), the control unit 31 determines whether there is a next manuscript before reading (S110). When it is determined that there is a next manuscript M (S110 "Yes"), S107 to S110 are repeated.

[0042] Then, when the control unit 31 determines that there is no next manuscript M (S110 "No"), it causes the manuscript images of each manuscript M stored in the image memory 26 to be formed on the respective recording papers (S111). After that, the process ends.

[0043] Also, when the paper deficiency determination unit 32 determines that there is a deficiency in the paper shape of the manuscript M (S108 "Yes") and the description information deficiency determination unit 33 determines that there is a deficiency in the description information of the manuscript M (S109 "Yes"), the determination unit 34 determines that there is a deficiency in the manuscript M (S112). When the control unit 31 is determined by the determination unit 34 that there is a deficiency in the manuscript M, it determines whether it stores an interruption instruction at deficiency in S104 or a continuation instruction at deficiency in S105 (S113).

[0044] Then, when the control unit 31 determines that it stores an interruption instruction at deficiency (S113 "Interruption"), it immediately stops the image reading unit 11 (S114) and displays a warning screen G2 as shown in FIG. 9 on the screen of the display unit 21 (S115).

[0045] On the warning screen G2 shown in Fig. 9, an error message E indicating that there is a missing original manuscript M is displayed, the image and page order of the original manuscript M with the missing part are displayed, and after correcting the missing part of the original manuscript M and resetting the original manuscript M to the image reading unit 11 again, a key K21 for selecting to read each remaining original manuscript M including the original manuscript M, and a key K22 for selecting to re-read all the original manuscripts M are displayed.

[0046] For example, the user views the error message E and the image of the original manuscript M with the missing part, corrects the missing part of the original manuscript M determined to have a missing part, resets the original manuscript M to the image reading unit 11 again, and then touches the key K21. When the control unit 31 detects a correction instruction corresponding to the key K21 through the operation unit 22 (S116 "correction"), it deletes the original manuscript image of the original manuscript M with the missing part stored in the image memory 26 in S107 (S117). Then, the control unit 31 resumes the operation of the image reading unit 11 (S118). After that, the process returns to S107, and the control unit 31 causes the image reading unit 11 to read the original manuscript M whose missing part has been corrected, and stores the original manuscript image of the original manuscript M in the image memory 26 (S107). After that, S108 - S110 are repeated.

[0047] Note that when S108 - S110 are repeated and the control unit 31 determines that there is no next original manuscript M (S110 "No"), the control unit 31 causes the original manuscript images of each original manuscript M stored in the image memory 26 to be formed on the respective recording papers by the image forming unit 12 (S111). After that, the process ends.

[0048] In this case, each time it is determined that there is a missing part in the original manuscript M, the original manuscript image of the original manuscript M with the missing part is deleted from the image memory 26, and the original manuscript image of the original manuscript M whose missing part has been corrected is stored in the image memory 26. Therefore, the images of each original manuscript M without missing parts will be formed on the recording paper.

[0049] In S116, the case where the user touches the key K22 after viewing the error message E and the image of the manuscript M with a defect will be described. When the control unit 31 receives a retry instruction corresponding to the key K22 via the operation unit 22 (S116 "Retry"), it deletes the manuscript images of all the manuscripts M read up to this point stored in the image memory 26 (S119), and causes the display unit 21 to display a defect determination suspension screen G3 as shown in the example in FIG. 10 (S120).

[0050] On the defect determination suspension screen G3, an error message E indicating that there is a defect in the manuscript M is displayed, the image of the manuscript M with this defect and the page order are displayed, and the determination of the paper shape and the defect of the described information of the manuscript M is turned off, and a key K31 for receiving a defect determination off instruction to retry the reading of all the manuscripts M, and a key K32 for receiving a defect determination on instruction to retry the reading of all the manuscripts M while maintaining the determination of the paper shape and the defect of the described information of the manuscript M on are displayed.

[0051] For example, the user views the error message E and the image of the manuscript M with a defect, corrects the defect of the manuscript M, sets all the manuscripts M again in the image reading unit 11, and touches the key K32. When the control unit 31 receives a defect determination on instruction corresponding to the key K32 (S121 "On"), it keeps the determination of the paper shape and the defect of the described information of the manuscript M on. After that, the process returns to the process from S107.

[0052] In this case, since the manuscript images of each read manuscript M are deleted from the image memory 26 and the image of the manuscript M with the defect corrected and the image data indicating each other image are stored in the image data, it can be expected that the images of each manuscript M without defects will be formed on the recording paper.

[0053] Also, the user views the error message E, the image of the manuscript M with missing parts, and the page order, sets all the manuscripts M again in the image reading unit 11, and touches the key K31. When the control unit 31 receives a missing determination off instruction corresponding to the touch operation on the key K31 (S121 "off"), after setting it to an off setting where it does not perform a missing determination of the paper shape and description information of the manuscript M, it resumes the operation of the image reading unit 11 (S122). The control unit 31 causes the image reading unit 11 to read all the manuscripts M, stores the manuscript images of each manuscript M in the image memory 26 (S123), and causes the image forming unit 12 to form the manuscript images of each manuscript M on the recording paper respectively (S124). After that, the process ends.

[0054] In this case, regardless of whether there is a missing part in all the manuscripts M, all the manuscripts M are read, and the images of all the manuscripts M are formed on the respective recording papers.

[0055] Also, in S113, when the control unit 31 determines that it has stored a continue instruction at the time of missing (S113 "continue"), the control unit 31 causes the image reading unit 11 to read all the remaining manuscripts M following the manuscript M in which the missing part has been determined, and stores the manuscript images of the read manuscripts M in the image memory 26 (S125). After the manuscript image of the said manuscript M is stored in the image memory 26, the control unit 31 displays a warning screen G4 as shown in FIG. 11 on the screen of the display unit 21 (S126).

[0056] On the warning screen G4, an error message E indicating that there is a missing part in the manuscript M is displayed, the image and page order of the manuscript M with the missing part are displayed, and the determination of the missing part of the paper shape and description information of the manuscript M is turned off, and a key K41 for receiving a manuscript rereading instruction to reread all the manuscripts M and a key K42 for receiving a print instruction to form the images of all the manuscripts M stored in the image memory 26 on the respective recording papers are displayed.

[0057] For example, the user views the error message E and the image of the manuscript M with a defect, sets all the manuscripts M again in the image reading unit 11, and touches the key K41. When the control unit 31 receives a manuscript rereading instruction corresponding to the key K41 via the operation unit 22 (S127 "redo"), it deletes all the manuscript images of each manuscript M stored in the image memory 26 (S128), resumes the operation of the image reading unit 11 (S129), causes the image reading unit 11 to read all the manuscripts M, and stores the manuscript images of each manuscript M in the image memory 26 (S130). Then, the control unit 31 causes the image forming unit 12 to form the manuscript images of each manuscript M on the recording paper (S131). After that, the process ends.

[0058] In this case, regardless of whether each of the manuscripts M has a defect, all the manuscripts M are read, and the images of all the manuscripts M are formed on the respective recording papers.

[0059] Also, the user views the error message E and the image of the manuscript M with a defect, and touches the key K42. When the control unit 31 receives a printing instruction corresponding to the key K42 via the operation unit 22 (S127 "end"), it causes the image forming unit 12 to form the manuscript images of each manuscript M stored in the image memory 26 on the respective recording papers (S132). After that, the process ends.

[0060] In this case, after the reading of the images of all the manuscripts M is completed, regardless of whether each of the manuscripts M has a defect, the images of all the manuscripts M are promptly formed on the respective recording papers.

[0061] As described above, in this embodiment, when it is determined whether there is a defect in the paper shape of the document M and whether there is a defect in the description information of the document M, if it is determined that there is a defect in the document M, for example, even if there is a defect in the document image due to a corner fold of the document M, if the defect does not cause any problem in the description content of the document M, it is not determined that there is a defect in the document M, and the document image of the document M is regarded as a valid image, enabling a highly convenient defect determination of the document image for the user. As a result, although the user can grasp the content of the document image, the trouble of having the user re-scan the document M due to the determination that there is a defect in the document M can be saved.

[0062] Also, in the above embodiment, as a method for detecting a defect in the paper shape, a method based on deep learning using a CNN is exemplified. However, the paper defect determination unit 32 generates a plurality of classification models that classify (determine) whether there is a defect or no defect in the paper shape of the document M based on different criteria by varying the combination of the input image data and the classification result data (correct answer data), and selects any one of the classification models according to an instruction received by a GUI operation by the user, and determines whether the paper shape of the document M is defective or not using this selected classification model. For example, the control unit 31 displays each key associated with each criterion related to the detection of the paper shape of the document M on the setting screen G1 of the defect detection mode shown in FIG. 4, detects a touch operation on any one of the keys through the touch panel 23, and sets the criterion associated with the detected key.

[0063] Similarly, regarding the detection of missing description information in the manuscript M, multiple classification models are generated by varying the combination of input image data and classification result data (correct answer data) and classifying (judging) the presence or absence of missing description information in the manuscript M according to different respective criteria. Then, any one of the classification models can be selected by the user's operation on the GUI, and it may be determined whether the description information of the manuscript M is missing or not using this selected classification model. For example, the control unit 31 displays on the setting screen G1 of the missing detection mode shown in FIG. 4 each key associated with each criterion related to the detection of the description information of the manuscript M, detects a touch operation on any key through the touch panel 23, and sets the criterion associated with this detected key.

[0064] Also, as described above, when the four sides of the image of the manuscript M are extracted and, at the corners of the image of the manuscript M, the inclined sides inclined with respect to the four sides are equal to or greater than a preset threshold value, and it is necessary to detect the presence of a missing paper shape of the manuscript M, the criterion for judging the presence or absence of a missing paper shape of the manuscript M can be changed by changing the threshold value. For example, the control unit 31 displays on the setting screen G1 of the missing detection mode shown in FIG. 4 each key associated with a plurality of threshold values, detects a touch operation on any key through the touch panel 23, and sets the threshold value associated with this detected key.

[0065] Also, in the above embodiment, one embodiment of the present invention has been described using an image forming apparatus (multifunction peripheral), but this is merely an example. One embodiment according to the present invention may be an image reading apparatus having a scanner function for reading an image of a manuscript without an image forming mechanism, or a personal computer having no image forming mechanism but having a scanner function.

[0066] Also, the configuration and processing of the above embodiment described with reference to FIGS. 1 to 11 are merely an example of the present invention, and the present invention is not intended to be limited to such configuration and processing.

Description of Reference Numerals

[0067] 10 Image forming apparatus 11 Image reading unit 12 Image forming unit 21 Display unit 22 Operation unit 23 Touch panel 26 Image memory 28 Storage unit 29 Control unit 31 Control section 32 Paper shortage determination unit 33 Missing description information determination unit 34 Determination unit

Claims

1. An image reading unit that reads an image of a document, A control unit that causes the image reading unit to sequentially read one document at a time, A paper defect determination unit that determines whether there is a defect in the paper shape of the read document based on the document image obtained by reading the document by the image reading unit, A description information defect determination unit that determines whether there is a defect in the description information of the read document based on the document image obtained by reading by the image reading unit, An image reading apparatus comprising: a determination unit that determines that there is a defect in the document when it is determined by the paper defect determination unit that there is a defect in the paper shape of the document and it is determined by the description information defect determination unit that there is a defect in the description information of the document.

2. The determination unit determines whether there is a defect in the document based on the document image each time a document image obtained by reading the document by the image reading unit is obtained for one document, The control unit stops reading the images of the subsequent documents by the image reading unit at the time when the determination unit determines that there is a defect in the document. The image reading apparatus according to claim 1.

3. The determination unit determines whether there is a defect in the document based on the document image each time a document image obtained by reading the document by the image reading unit is obtained for one document, The control unit continues to read the images of the subsequent documents by the image reading unit when the determination unit determines that there is a defect in the document, and stops reading the images of the documents by the image reading unit after all the documents have been read. The image reading apparatus according to claim 1.

4. The paper defect determination unit has a plurality of different criteria for determining whether there is a defect in the paper shape of the document, and when any one of the criteria is selected, determines whether there is a defect in the paper shape of the document according to the selected criterion. The image reading apparatus according to claim 1.

5. The description information defect determination unit has a plurality of different criteria for determining whether there is a defect in the description information of the document, and when any one of the criteria is selected, determines whether there is a defect in the description information of the document according to the selected criterion. The image reading apparatus according to claim 1.

Images