Image processing system, image processing method, and program

Abstract

To provide an image processing system capable of reducing the capacity of an image to transmit to a server and shortening a time for transmission.SOLUTION: An image processing system 1 comprises: an image reading device 2 for reading a document; and a virtual server 4 for executing character recognition. The image reading device 2 comprises an image acquisition unit 201, an image cut-out unit 204, and an image transmission unit 206. The image acquisition unit 201 acquires a read image from the document. The image cut-out unit 204 cuts out a cut-out image needed for character recognition from the read image acquired by the image acquisition unit 201. The image transmission unit 206 transmits the cut-out image cut out by the image cut-out unit 204 to the virtual server 4. The virtual server 4 comprises a character recognition unit 404 for executing character recognition for the cut out image transmitted from the image reading device 2.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present invention relates to an image processing system, an image processing method, and a program. [Background technology]

[0002] Conventionally, at financial institution branches, operators such as bank clerks have used image reading devices such as OCR (Optical Character Readers) that use scanners to read characters written on documents and the like.

[0003] When using the OCR function in an image reading device such as an OCR, the form image is read by an image reading device such as an OHR (Over Head Reader) installed locally. The read form image must then be transferred to an image processing server via a device server connected to the network. In addition, images used in financial transactions must be saved as high-quality images in order to leave a trail. For this reason, images sent (transferred) from the image reading device to the image processing server must be of high quality.

[0004] In view of this, a form recognition system has been disclosed in which a server and a system terminal equipped with an image reading device that reads handwritten characters are connected to each other via a communication line network so that they can communicate with each other (see, for example, Patent Document 1).The server then recognizes the handwritten characters in the image data of the form read by the image reading device. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2019-204417 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, when a form image is transmitted (transferred) from a system terminal equipped with an image reading device to a server, the image size becomes huge because the image needs to be of high image quality, and transmission (transfer) takes a long time. Also, when form images are transmitted from multiple image reading devices to an image processing server at the same time, network traffic occurs, and there is a risk that financial transactions will not proceed smoothly.

[0007] SUMMARY OF THE PRESENT DISCLOSURE In view of the above problems, an object of the present invention is to provide an image processing system capable of reducing the volume of images transmitted to a server and shortening the transmission time. [Means for solving the problem]

[0008] The present invention is an image processing system comprising an image reading device that reads a form and an image processing server that performs character recognition, wherein the image reading device comprises an image acquisition unit that acquires a read image from the form, an image cutting unit that cuts out a cut-out image required for character recognition from the read image acquired by the image acquisition unit, and an image transmission unit that transmits the cut-out image cut out by the image cutting unit to the image processing server, and the image processing server comprises a character recognition unit that performs character recognition on the cut-out image transmitted from the image reading device.

[0009] The present invention also provides an image processing method executed by an image processing system having an image reading device that reads a form and an image processing server that performs character recognition, wherein the image reading device acquires a read image from the form, cuts out a cut-out image required for character recognition from the acquired read image, transmits the cut-out image to the image processing server, and the image processing server performs character recognition on the cut-out image transmitted from the image reading device.

[0010] The present invention also relates to a program executed by an image processing system having an image reading device that reads a form and an image processing server that performs character recognition, wherein the image reading device acquires a read image from the form, cuts out a cut-out image required for character recognition from the acquired read image, transmits the cut-out image to the image processing server, and the image processing server performs character recognition on the cut-out image transmitted from the image reading device. Effect of the Invention

[0011] According to one aspect of the present invention, it is possible to reduce the volume of images to be transmitted to a server and shorten the transmission time. [Brief description of the drawings]

[0012] [Figure 1] 1 is a system configuration diagram of an image processing system including an image reading device and a virtual server according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a hardware configuration diagram of a virtual server according to the present embodiment. [Diagram 3] FIG. 2 is a diagram showing an example of functional blocks of an image processing system 1 for image character recognition processing according to the present embodiment. [Figure 4] FIG. 4 is a diagram showing an example of a read image acquired by an image acquisition unit; [Diagram 5] FIG. 2 is a diagram showing an example of a cut-out image according to the embodiment; [Figure 6] FIG. 4 is a diagram showing an example of a template image according to the embodiment; [Figure 7] FIG. 2 is a diagram showing an example of a composite image according to the embodiment; [Figure 8] 4 is a sequence chart showing an example of an image character recognition process executed by the image processing system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Fig. 1 is a system configuration diagram of an image processing system 1 including an image reading device 2 and a virtual server 4 according to the present embodiment.

[0014] 1, the image processing system 1 includes an image reading device 2, a device 3, and a virtual server 4. The image reading device 2 and the virtual server 4 are connected to each other so as to be able to communicate with each other.

[0015] Here, the image processing system 1 of this embodiment is installed in, for example, a financial institution, a public institution, etc. Also, the image processing system 1 of this embodiment can be configured with an image reading device 2 and a virtual server 4 as functions built into a mobile communication terminal.

[0016] The image reading device 2 includes a base 21, a camera 22, and a control unit 23. The image reading device 2 is, for example, a stand-type scanner such as an OHR (Over Head Reader), and optically reads a form placed on the base 21 with the camera 22 to obtain a read image. The control unit 23 cuts out a cutout image required for character recognition from the obtained read image, and sends it to the virtual server 4 via the device 3. The form to be read is, for example, a transfer request form, a deposit / withdrawal slip, or other document that requires various handwritten input. In FIG. 1, one image reading device 2 is installed, but this is not limited thereto, and two or more image reading devices may be installed.

[0017] The device 3 is a connection interface (I / F) that connects the image reading device 2 and the virtual server 4. The image reading device 2 and the device 3 are connected by a Universal Serial Bus (USB). The device 3 and the virtual server 4 are also connected by a Local Area Network (LAN) through a network N. The device 3 transfers the cut-out image sent from the image reading device 2 to the virtual server 4 through the network N. In FIG. 1, one device 3 is installed according to the number of image reading devices 2, but this is not limited thereto, and two or more devices may be installed.

[0018] The virtual server 4 has the functions of a computer such as a workstation, a server, or a personal computer, and executes image character recognition processing based on an image processing method, which will be described later. The virtual server 4 is an example of an image processing server. In FIG. 1, one virtual server 4 is installed, but this is not limited thereto, and two or more virtual servers may be installed.

[0019] FIG. 2 is a hardware configuration diagram of the virtual server 4 according to the present embodiment.

[0020] 2, the virtual server 4 is configured by a control unit 41, an input device 42, an output device 43, a storage unit 44, and an interface (I / F) 45, all of which are connected to a bus 46. The virtual server 4 may also include other hardware configurations not shown in FIG.

[0021] The control unit 41 comprehensively controls the entire virtual server 4, and includes a CPU (Central Processing Unit) that reads a program and executes image character recognition processing.

[0022] The input device 42 is, for example, a keyboard, a joystick, a light pen, a mouse, a touch pad, a touch panel, a trackball, or the like, and inputs various types of data and signals.

[0023] The output device 43 is, for example, various displays such as an LCD (Liquid Crystal Display), a printer, etc., and outputs images and other information.

[0024] The storage unit 44 has a RAM (Random Access Memory), a ROM (Read Only Memory), an HDD (Hard Disk Drive), an SSD (Solid State Drive), etc. The storage unit 44 stores a program for the virtual server 4 to execute image character recognition processing, a control program for controlling and executing each function of the virtual server 4, screen data, etc. The storage unit 44 provides a working area when the virtual server 4 executes a program. The storage unit 44 may also store other data such as definition body information and composite images described below.

[0025] The interface 45 is a unit for connecting to the image reading device 2 and external devices (not shown), such as a serial interface such as a Universal Serial Bus (USB) or a parallel interface such as a Local Area Network (LAN).

[0026] The image character recognition process according to the present embodiment is executed by the image reading device 2 and the virtual server 4 described above.

[0027] Fig. 3 is a diagram showing an example of functional blocks of the image processing system 1 relating to the image character recognition processing of the present embodiment. As shown in Fig. 3, the control unit 23 of the image reading device 2 includes an image acquisition unit 201, a symbol acquisition unit 202, a symbol transmission unit 203, an image cutout unit 204, an encryption unit 205, and an image transmission unit 206. The control unit 23 may include other components not shown in Fig. 3.

[0028] The image acquisition unit 201 acquires a read image by photographing and reading a document placed on a base 21 with a camera 22 .

[0029] Fig. 4 is a diagram showing an example of a scanned image 500 acquired by the image acquisition unit 201. The scanned image 500 includes a print area, a character area, and an identification area. The print area is an area of ​​items that are printed in advance on the document. The scanned image 500 in Fig. 4 includes items such as the bank name, branch name, deposit type, account number, amount, beneficiary last name, beneficiary first name, beneficiary, beneficiary address, client last name, client name, client, and client address as the print area.

[0030] The character area is an area where a user who fills out the form has handwritten input using a writing implement such as a ballpoint pen, pencil, etc. The scanned image 500 in Fig. 4 includes areas 51 to 56 as character areas.

[0031] The identification area is an area where a document ID for identifying the type of document is printed. The scanned image 500 in FIG. 4 includes an area 57 as an identification area. The document ID is an example of an identification symbol for identifying the type of document. The document ID corresponds to the type of document. For example, in the embodiment in FIG. 4, the document ID "0101" is set in correspondence with the document type "transfer request form." Although not shown, different document IDs are set for other documents such as a deposit application form, a refund request form, a currency exchange request form, and a refund request form in correspondence with the type of document. Therefore, the type of document can be identified by recognizing the document ID. The coordinate position of the identification area is set to a common position regardless of the type of document.

[0032] The symbol acquisition unit 202 acquires a form ID for identifying the type of form from the scanned image 500 acquired by the image acquisition unit 201. In the embodiment of Fig. 4, "0101" is printed as the form ID in the area 57. Therefore, the symbol acquisition unit 202 performs image analysis of the characters printed in the area 57 of the scanned image 500 shown in Fig. 4 to acquire the form ID of "0101".

[0033] The symbol sending unit 203 sends the form ID of the form acquired by the symbol acquiring unit 202 to the virtual server 4. In this case, the symbol sending unit 203 sends the form ID to the virtual server 4 through the device 3. In this embodiment, the symbol sending unit 203 sends “0101” to the virtual server 4 as the form ID.

[0034] The image cutting unit 204 cuts out a cut-out image required for character recognition of the form from the scanned image 500 acquired by the image acquisition unit 201. In this case, the image cutting unit 204 cuts out a cut-out image of a character region corresponding to the type of the form. Specifically, the image cutting unit 204 cuts out a cut-out image of a character region corresponding to the type of the form based on the definition body information transmitted from the virtual server 4. The definition body information is information transmitted from the virtual server 4 based on the form ID transmitted from the image reading device 2.

[0035] The definition information includes information on the coordinates where the character area exists. In the embodiment shown in FIG. 4, the coordinates of areas 51 to 56, which are character areas corresponding to the form ID "0101", are defined. The image cutting unit 204 can determine the area of ​​the cut-out image to be cut out by referring to the definition information corresponding to the form ID "0101". Therefore, in the embodiment of FIG. 4, the image cutting unit 204 cuts out areas 51 to 55 corresponding to the form ID "0101" as cut-out images.

[0036] Fig. 5 is a diagram showing an example of a cutout image 600 according to this embodiment. In the embodiment of Fig. 5, cutout image 600 includes areas 61 to 66. Cutout image 600 is an area corresponding to areas 51 to 56 which correspond to the character area in Fig. 4. In other words, areas 51 to 56 in Fig. 4 correspond to areas 61 to 66 in Fig. 5.

[0037] When the cut-out image 600 is transferred to the virtual server 4, it is necessary to ensure reliability that the transferred image is not forged by being replaced with forged information. Therefore, the encryption unit 205 encrypts the cut-out image 600 cut out by the image cut-out unit 204. As shown in FIG. 5, when a plurality of cut-out images 600 are transferred to the virtual server 4, encryption using the plurality of cut-out images 600 is possible. For example, the encryption unit 205 encrypts the cut-out image 600 based on the items of each cut-out image 600 cut out by the image cut-out unit 204. As an example, an embodiment using a public key system will be described. For example, the encryption unit 205 performs encryption using the following method.

[0038] 1. Generate a hash value for the cropped image of area 61 corresponding to the bank name. 2. The cut-out image of area 61 corresponding to the bank name is encrypted with the public key. 3. Generate a hash value for the cropped image of area 62 corresponding to the branch name. 4. The cut-out image of the area 62 corresponding to the branch name is encrypted using a hash value generated based on the bank name as a key. 5. Repeat steps 3 and 4 above for other items (deposit type, account number, amount, beneficiary last name, beneficiary first name, beneficiary, beneficiary address, requester last name, requester first name, requester, requester address, etc.). By applying the above method, the cut-out image 600 can be encrypted.

[0039] The image sending unit 206 sends the cut-out image 600 cut out by the image cutting unit 204 to the virtual server 4. In this case, the image sending unit 206 sends the cut-out image 600 to the virtual server 4 through the device 3. If the cut-out image 600 has been encrypted by the encryption unit 205, the image sending unit 206 sends the encrypted cut-out image 600 to the virtual server 4.

[0040] For example, in this embodiment, the read image 500 acquired from the image reading device 2 has an image capacity of 1.6 MB, and the cut image 600 cut out by the image cutout unit 204 has an image capacity of 0.6 MB. In this regard, the image transmission unit 206 of this embodiment transfers only the cut image 600 cut out by the image cutout unit 204, out of the read image 500 acquired from the image reading device 2, to the virtual server 4. Therefore, it is possible to reduce the capacity of the transferred image and shorten the transfer time.

[0041] 3, the control unit 41 of the virtual server 4 includes a definition body search unit 401, a definition body transmission unit 402, a decoding unit 403, a character recognition unit 404, a composite image generation unit 405, and a composite image storage unit 406. The control unit 41 may include other components not shown in FIG. 3. The storage unit 44 of the virtual server 4 includes a definition body storage unit 441, a template storage unit 442, and a composite image storage unit 443. The storage unit 44 may store data other than the definition body storage unit 441, the template storage unit 442, and the composite image storage unit 443.

[0042] At least one definition body for a form is stored in the definition body storage unit 441. The definition body includes the coordinates of a character area for which character recognition corresponding to the type of form is to be performed. Since the definition body is stored for each form, different definition bodies are stored for the number of types of forms.

[0043] Form templates are stored in the template storage unit 442. In this embodiment, a template is a model of a form in which areas other than the text area are filled out.

[0044] In the composite image storage unit 443, a composite image, which will be described later and which is generated by the composite image generation unit 405, is stored by the composite image saving unit 406. This makes it possible to save a high-quality image for the purpose of leaving a trail of the image used in the financial transaction, without transmitting the entire scanned image 500, which has a large image capacity, from the image reading device 2 to the virtual server 4. This makes it possible to reduce the capacity of the image to be transmitted to the virtual server 4, and shorten the transmission time. Furthermore, since the capacity of the image can be reduced, the cost of the device required for storage can be suppressed.

[0045] The definition body storage unit 441 stores at least one definition body information for a form. The definition body includes the coordinates of a character area for which character recognition corresponding to the type of form is to be performed. Since the definition body is stored for each form, different definition body information is stored for each type of form. Specifically, the definition body storage unit 441 stores definition body information linked to each form ID. This makes it possible to easily search for definition body information for a corresponding type of form using the form ID as a key.

[0046] The definition body search unit 401 searches the definition body of the form corresponding to the identification symbol transmitted from the image reading device 2 from the definition body storage unit 441 of the storage unit 44 .

[0047] The definition body transmission unit 402 transmits the definition body information searched by the definition body search unit 401 to the image reading device 2. The definition body information includes information on the coordinates where the character region exists. Different information exists in the definition body information corresponding to each form ID.

[0048] If the cut-out image 600 transmitted from the image reading device 2 is encrypted, the decryption unit 403 decrypts the encrypted cut-out image 600. As shown in FIG. 5, if the cut-out image 600 transferred to the virtual server 4 is composed of a plurality of regions, the decryption is performed using the cut-out images 600 of the plurality of regions. For example, the decryption unit 403 decrypts the cut-out image 600 based on the items of each region that composes the cut-out image 600 cut out by the image cut-out unit 204. As an example, an embodiment using a public key system will be described. For example, the decryption unit 403 performs decryption using the following method.

[0049] 6. The cropped image of area 61 corresponding to the bank name is decrypted with the private key. 7. Generate a hash value for the cropped image of region 61 corresponding to the bank name. 8. The cut-out image of area 62 corresponding to the branch name is decrypted using a hash value generated based on the bank name. 9. Generate a hash value for region 62 corresponding to the branch name. 10. Repeat steps 8 and 9 for other items (deposit type, account number, amount, beneficiary last name, beneficiary first name, beneficiary, beneficiary address, client last name, client first name, client, client address, etc.) By applying the above method, the cut-out image 600 can be decoded.

[0050] The character recognition unit 404 performs character recognition on the cut-out image 600 transmitted from the image reading device 2. A known OCR (Optical Character Reader) technique can be applied for the character recognition. If the cut-out image 600 transmitted from the image reading device 2 is encrypted, the character recognition unit 404 performs character recognition on the cut-out image 600 decrypted by the decryption unit 403.

[0051] The composite image generating unit 405 generates a composite image by combining the cutout image 600 transmitted from the image reading device 2 with a template image corresponding to the type of the form. Fig. 6 is a diagram showing an example of a template image 700 according to this embodiment. Fig. 7 is a diagram showing an example of a composite image 800 according to this embodiment.

[0052] Template image 700 is an image having blank areas obtained by deleting the character areas shown in FIG. 4 from the document. In the embodiment of FIG. 6, areas 71 to 76 are formed which correspond to blank areas corresponding to areas 51 to 56 which correspond to the character areas. Since areas 71 to 76 are blank areas, cut-out image 600 can be superimposed on template image 700. By superimposing areas 51 to 56 which correspond to the character areas on areas 71 to 76 which correspond to the blank areas of template image 700 to synthesize composite image 800, it is possible to generate composite image 800 which is the same as original scanned image 500 acquired by image reading device 2, as shown in FIG. 7.

[0053] The composite image saving unit 406 saves the composite image 800 generated by the composite image generating unit 405 in the composite image storage unit 443. This allows the high-quality image used in the financial transaction to be saved for the purpose of leaving a trail.

[0054] Next, a description will be given of the image character recognition processing executed by the image processing system 1. Fig. 8 is a sequence chart showing an example of the image character recognition processing executed by the image processing system 1. The image character recognition processing according to this embodiment is executed by the image reading device 2 and the virtual server 4 described using Fig. 3 according to an image character recognition processing program.

[0055] The image character recognition processing program is executed according to the image character recognition processing flow shown in FIG.

[0056] First, the control unit 23 of the image reading device 2 captures an image of the document placed on the base 21 with the camera 22 (step S11). The control unit 23 acquires the scanned image 500 captured and read in step S11 (step S12).

[0057] The control unit 23 acquires the form ID written in the area 57 corresponding to the identification area from the scanned image 500 of the form acquired in step S12 (step S13). Then, the control unit 23 transmits the form ID acquired in step S13 to the virtual server 4 (step S14).

[0058] The control unit 41 of the virtual server 4 receives the form ID transmitted from the image reading device 2 (step S15). The control unit 41 searches for definition body information corresponding to the form ID received in step S15 by referring to the definition body storage unit 441 (step S16). The control unit 41 transmits the definition body information searched for in step S16 to the image reading device 2 (step S17).

[0059] The control unit 23 of the image reading device 2 cuts out the cut-out images 600 from the read image 500 acquired in step S12 based on the definition information transmitted from the virtual server 4 (step S19). The control unit 23 encrypts the cut-out images 600 based on the items of each cut-out image 600 cut out in step S19 (step S20). The control unit 23 transmits the cut-out images 600 encrypted in step S20 to the virtual server 4 (step S21).

[0060] The control unit 41 of the virtual server 4 receives the cut-out image 600 transmitted from the image reading device 2 (step S22). The control unit 41 decodes the cut-out image 600 received in step S22 (step S23).

[0061] The control unit 41 performs character recognition by OCR on the characters included in the cut-out image 600 decoded in step S23 (step S24).

[0062] The control unit 41 searches the template storage unit 442 for the template image 700 corresponding to the form ID received in step S15. Then, the control unit 41 generates a composite image 800 by combining the cut-out image 600 transmitted from the image reading device 2 and the template image 700 corresponding to the form ID (step S25).

[0063] Control unit 41 stores composite image 800 created in step S24 in composite image storage unit 443 (step S26). When this process ends, the image character recognition process ends.

[0064] According to the above embodiment, the image cutting unit 204 cuts out a cut image required for character recognition of a form from the scanned image 500 acquired by the image acquisition unit 201. Then, the image transmission unit 206 transmits only the cut image 600 cut out by the image cutting unit 204 to the virtual server 4. Therefore, it is possible to reduce the capacity of the image transmitted from the image reading device 2 to the virtual server 4 and shorten the transmission time.

[0065] Furthermore, the composite image saving unit 406 saves the composite image generated by the composite image generating unit 405 in the composite image storage unit 443. This makes it possible to save a high-quality image for the purpose of leaving a trail of the image used in the financial transaction, without transmitting the entire scanned image 500, which has a large image capacity, from the image reading device 2 to the virtual server 4. This makes it possible to reduce the capacity of the image to be transmitted to the virtual server 4 and shorten the transmission time. Furthermore, since the capacity of the image can be reduced, the cost of the device required for storage can be suppressed.

[0066] The present invention is not limited to the above-described embodiment as it is, and the components can be modified and embodied within the scope of the gist of the invention at the implementation stage. In addition, various inventions can be formed by appropriate combinations of multiple components disclosed in the above embodiments. For example, all components shown in the embodiments may be appropriately combined. Furthermore, components from different embodiments may be appropriately combined. Various modifications and applications are possible within the scope of the invention. [Explanation of symbols]

[0067] 1: Image processing system 2: Image reader 3: Device 4: Virtual Server 21: Pedestal 22: Camera 23: Control section 41: Control section 42: Input device 43: Output device 44: Storage section 45: Interface 46: Bus 51~56: Area 57 :Area 61~66: area 71~76: area 201: Image acquisition unit 202: Symbol acquisition section 203: Symbol transmission unit 204: Image cutout section 205: Encryption section 206: Image transmission unit 401: Definition body search section 402: Definition body transmission section 403: Decoding section 404:Character recognition section 405: Composite image generation unit 406: Composite image storage unit 441: Definition field storage section 442: Template storage section 443: Synthetic image storage unit 500: Scanned image 600: Cutout image 700: Template image 800: Composite image N: Network

Claims

1. An image processing system including an image reading device that reads a form and an image processing server that performs character recognition, The image reading device includes: an image acquisition unit for acquiring a scanned image from the document; an image cutout unit that cuts out a first cutout image and a second cutout image necessary for character recognition from the read image acquired by the image acquisition unit; an encryption unit that encrypts the first cut-out image cut out by the image cut-out unit, and encrypts the second cut-out image cut out by the image cut-out unit using a hash value of the first cut-out image; an image transmission unit that transmits the first cut-out image and the second cut-out image encrypted by the encryption unit to the image processing server, The image processing server includes: a decoding unit that decrypts the encrypted first cut-out image transmitted from the image reading device, and decrypts the encrypted second cut-out image transmitted from the image reading device using a hash value of the decrypted first cut-out image; a character recognition unit that performs character recognition on the first cut-out image and the second cut-out image decoded by the decoding unit.

1. An image processing system comprising:

2. 2. The image processing system according to claim 1, the image cutout unit cuts out the first cutout image of a first character region corresponding to the type of the form and the second cutout image of a second character region corresponding to the type of the form; 1. An image processing system comprising:

3. 3. The image processing system according to claim 2, The image reading device includes: a symbol acquisition unit that acquires an identification symbol that identifies the type of the document from the scanned image acquired by the image acquisition unit; a symbol transmission unit that transmits the identification symbol to the image processing server, The image processing server includes: a definition body search unit that searches for definition body information of the form corresponding to the identification code transmitted from the image reading device; A definition body transmission unit that transmits the definition body information inspected by the definition body search unit to the image reading device, The image cutout unit cuts out the first cutout image of the first character region corresponding to the type of the form and the second cutout image of the second character region corresponding to the type of the form based on the definition information transmitted from the image processing server.

1. An image processing system comprising:

4. The image processing system according to any one of claims 1 to 3, The image processing server includes: a composite image generating unit that generates a composite image by combining the first cut-out image and the second cut-out image decoded by the decoding unit with a template image corresponding to the type of the form; a composite image storage unit that stores the composite image generated by the composite image generation unit.

1. An image processing system comprising:

5. An image processing method executed by an image processing system including an image reading device that reads a form and an image processing server that performs character recognition, comprising: The image reading device obtains a read image from the document, a first cut-out image and a second cut-out image required for character recognition are cut out from the acquired read image; encrypting the first cropped image; encrypting the second cropped image using a hash value of the first cropped image; Sending the encrypted first cropped image and the encrypted second cropped image to the image processing server; The image processing server decrypts the encrypted first cut-out image transmitted from the image reading device, decrypting the encrypted second cut-out image transmitted from the image reading device using a hash value of the decrypted first cut-out image; Character recognition is performed on the decoded first cut-out image and the decoded second cut-out image.

13. An image processing method comprising:

6. A program executed by an image processing system including an image reading device that reads a form and an image processing server that performs character recognition, The image reading device obtains a read image from the document, a first cut-out image and a second cut-out image required for character recognition are cut out from the acquired read image; encrypting the first cropped image; encrypting the second cropped image using a hash value of the first cropped image; Sending the encrypted first cropped image and the encrypted second cropped image to the image processing server; The image processing server decrypts the encrypted first cut-out image transmitted from the image reading device, decrypting the encrypted second cut-out image transmitted from the image reading device using a hash value of the decrypted first cut-out image; Performing character recognition on the decoded first cut-out image and the decoded second cut-out image. A program characterized by:

Images