JP6492622B2 - Character image processing system, information processing apparatus, and control program for information processing apparatus - Google Patents

Description

The present invention is a character image processing system, information processing apparatus, and a control program of the information processing apparatus. More particularly, the present invention relates to an information processing apparatus and the OCR (Optical Character Recognition) text image processing system comprising a device, information processing apparatus, and a control program of the information processing apparatus.

  An MFP (Multifunction Peripheral), which is one of image forming apparatuses, includes a scanner function, a facsimile function, a copying function, a printer function, a data communication function, and a server function.

  Some recent MFPs have a function of creating a searchable PDF (Portable Document Format) (registered trademark) using scanned image data. The searchable PDF is a PDF file obtained by converting characters included in a scanned document image into text data by OCR processing and combining the text data with the document image. The searchable PDF includes a base layer and a transparent layer above the base layer. The base layer includes image data such as JPEG formation. The transparent layer is composed of text data obtained by OCR processing.

  Searchable PDF enables character (keyword) search in a document. It is also possible to copy and paste characters in a document into another digital document. Therefore, it is not necessary to type characters in a paper document in order to digitize the document.

  In addition, the technique regarding creation of searchable PDF is disclosed by the following patent document 1, etc., for example.

  In order to create a searchable PDF, OCR processing is required as described above. OCR processing places a heavy load on the CPU (Central Processing Unit). For this reason, when OCR processing is performed in the MFP, there is a possibility that other operations of the MFP (for example, copy operation, scan operation, print operation, facsimile transmission / reception, etc.) may be hindered.

  Among the processes for creating a searchable PDF, the MFP performs a process for creating scanned image data, a process for determining a character area, and a process for creating a PDF image by hardware. Therefore, the MFP can perform these processes instantaneously. On the other hand, the MFP performs OCR processing by software. For this reason, the MFP tends to occupy the CPU for a long time for OCR processing.

  A technique for avoiding the occupation of the CPU for a long time by OCR processing is disclosed in, for example, Patent Document 2 below. In the technique of Patent Document 2 below, a technique for managing the calculation load by OCR processing and controlling the OCR processing when the MFP is operating is disclosed. However, this technique lowers the priority of the OCR process in order to prioritize the operation of the MFP other than the OCR process. Therefore, there is a problem that it takes time to create a searchable PDF.

  Therefore, a technique for using an OCR site of an external server on the Internet has been proposed. This technique is disclosed in Patent Document 3 below, for example. The external server releases the storage area to the individual as one of the services provided to the individual. The OCR site transfers image data received from a client (such as an MFP, a PC (Personal Computer), or a portable terminal) to an external server. The external server performs OCR processing on the transferred image data and stores the obtained text data. The client can browse and acquire text data. By using this OCR site, it is possible to avoid long-time occupation of the CPU due to OCR processing.

JP 2012-73749 A JP 2013-161268 A JP 2010-9131 A

  However, when performing OCR processing using an external OCR site, there is a problem that confidential information is likely to leak. The leakage of confidential information is caused by unauthorized access by a third party to the OCR site.

  Usually, at the OCR site, the ID and password for login are managed using SSL (Secure Sockets Layer) or the like. However, if a third party steals these IDs or passwords by means of packet sniffing or the like, the third party can view or download personal information and manuscript images stored on an external server of the OCR site. It becomes possible to do. Also, a third party can view or download data transmitted / received between an external server of the OCR site and a personal terminal.

The present invention is intended to solve the above problems, its object is to provide a character imaging system that can suppress the leakage of confidential information, information processing apparatus, and a control program of an information processing apparatus It is.

A character image processing system according to one aspect of the present invention includes a first information processing unit and a second image processing unit having an OCR function capable of communicating with the first information processing unit via a network. In the processing system, the first information processing unit includes an image block creating unit that divides a character area in the image data into a plurality of image blocks, an arrangement order changing unit that changes an arrangement order of the plurality of image blocks, Inserting means for inserting a connecting image between each of the plurality of image blocks after the arrangement order is changed by the arrangement order changing means, and a plurality of image blocks after the arrangement order is changed by the arrangement order changing means A first transmission unit configured to transmit an encrypted image generated based on the encrypted image including a plurality of image blocks into which the connection image is inserted to the second information processing unit; Information processing Performs OCR processing on the encrypted image, thereby generating OCR processing means for creating first text data and second OCR data including the first text data to the first information processing unit. The first information processing unit further decomposes the post-OCR data into a plurality of character strings based on the concatenated image, and deletes characters corresponding to the concatenated image from the post-OCR processed data. Thus, a creation means for creating the second text data based on the post-OCR data and a pasting means for pasting the second text data to the corresponding positions of the character areas in the image data are included.

  Preferably, in the character image processing system, the first information processing unit further includes character region specifying means for specifying a character region in the image data and specifying the coordinates of the character region in the image data. The second text data is pasted based on the coordinates.

  Preferably, in the character image processing system, the pasting unit pastes the second text data at a position corresponding to the character region in the transparent layer of the image data.

  Preferably, in the character image processing system, the first information processing unit further includes a level receiving unit that receives a security level setting, and the image block creating unit is determined according to the level received by the level receiving unit. A character area is divided into a plurality of image blocks of a size.

  Preferably, in the above character image processing system, the second information processing unit includes a first OCR device and a second OCR device different from the first OCR device, and the first transmission means includes an encryption unit. The first portion of the encrypted image is transmitted to the first OCR device, and the second portion of the encrypted image that is different from the first portion is transmitted to the second OCR device.

  Preferably, in the character image processing system, the first information processing unit further includes an image reading unit that creates image data by reading an image of a document.

  Preferably, in the character image processing system, the image block creation means is a distribution of the number of white pixels accumulated in a first direction which is the direction of one side in a rectangular character region, and the first direction A first distribution extracting means for extracting a distribution along a second direction perpendicular to the second distribution, and a distribution of the number of white pixels existing in the second direction in a rectangular character region, A character area in the image data is divided at a position determined based on the distribution extracted by the second distribution extracting means for extracting the distribution along the direction 1 and each of the first and second distribution extracting means. And a dividing unit for creating a plurality of image blocks.

  Preferably, in the above character image processing system, the dividing unit divides the character area between the line specified by the line specifying unit and the line specifying unit that specifies the line spacing based on the distribution extracted by the first distribution extracting unit. The second distribution extracting means includes white line existing in the second direction for each of the plurality of lines divided by the line dividing means. A distribution of the number of pixels integrated, wherein a distribution along the first direction is extracted, and the dividing unit specifies a gap position of the character based on the distribution extracted by the second distribution extracting unit. Identifying means; boundary determining means for determining the boundary position based on the gap position specified by the gap specifying means; and column direction dividing means for dividing each of the plurality of rows at the boundary position determined by the boundary determining means. And further includes.

  Preferably, in the above character image processing system, the boundary determining unit determines, as the boundary position, a gap position in which the distance from another adjacent gap position is equal to or greater than a threshold among the gap positions specified by the gap specifying unit.

In the character image processing system, preferably, the connection image is an image whose character recognition result is known and is held in advance by the first information processing unit .

  In the character image processing system, preferably, the connection image is an image of a symbol that is not a character.

  Preferably, in the character image processing system, the first information processing unit changes the order of each of the plurality of image blocks before the arrangement order is changed by the arrangement order changing unit and the arrangement order by the arrangement order changing unit. The image processing apparatus further includes array information holding means for holding relation information indicating a relation with the order of each of the plurality of image blocks.

  In the character image processing system, the network is preferably the Internet.

  Preferably, in the character image processing system, the first information processing unit includes an image forming apparatus capable of optically reading a document.

  Preferably, in the character image processing system, the first information processing unit further includes a terminal separate from the image forming apparatus, and the first transmission unit transmits the encrypted image from the terminal to the second information processing unit. Send.

  Preferably, in the character image processing system, the first information processing unit generates an encrypted image based on the optically read image data.

An information processing apparatus according to another aspect of the present invention is an information processing apparatus that communicates with an OCR apparatus, and includes an image block creating unit that divides a character region in image data into a plurality of image blocks, and a plurality of image blocks. An arrangement order changing means for changing the arrangement order, an insertion means for inserting a linking image between each of the plurality of image blocks after the arrangement order is changed by the arrangement order changing means, and an arrangement by the arrangement order changing means A transmitting means for transmitting an encrypted image created based on a plurality of image blocks after changing the order and including a plurality of image blocks into which concatenated images are inserted, to the OCR device; the OCR post-data including the first text data created by performing the OCR process on the basis of an image, receiving means for receiving from the OCR device, based on the consolidated image O Decomposes the R post data to a plurality of strings by removing the character that corresponds to the connection image from the OCR-processed data, and generating means for generating second text data based on OCR after data, image data Pasting means for pasting the second text data to the corresponding positions of the character areas.

An information processing device control program according to still another aspect of the present invention is an information processing device control program for communicating with an OCR device, and an image block creation step of dividing a character region in image data into a plurality of image blocks. An arrangement order changing step for changing the arrangement order of the plurality of image blocks, and an insertion step for inserting a linking image between each of the plurality of image blocks after the arrangement order is changed in the arrangement order changing step; An encrypted image created on the basis of a plurality of image blocks after the arrangement order is changed in the arrangement order changing step and including the plurality of image blocks with the concatenated images inserted therein is sent to the OCR device A post-OCR data including a transmission step for transmission and first text data created by performing OCR processing based on the encrypted image; And a receiving step of receiving from the OCR device, decomposing the OCR post data to a plurality of character strings on the basis of the consolidated image, by removing the character that corresponds to the connection image from the OCR-processed data, OCR after data And a pasting step for pasting the second text data to the corresponding positions of the character areas in the image data.

According to the present invention, it is possible to provide a character imaging system that can suppress the leakage of confidential information, information processing apparatus, and a control program of the information processing apparatus.

It is a block diagram which shows notionally the structure of the document image processing system in the 1st Embodiment of this invention. It is a sequence diagram which shows the outline | summary of operation | movement of the character image processing system in the 1st Embodiment of this invention. In the 1st Embodiment of this invention, it is a figure which shows typically the screen SR displayed on the operation panel of the tablet terminal. FIG. 4 is a diagram schematically showing character areas L1, L2, and L3 included in read image data IM in the first embodiment of the present invention. It is a figure which shows typically distribution of each white pixel of the x direction and the y direction in the image of the character area L1 in the 1st Embodiment of this invention. In the 1st Embodiment of this invention, it is a figure which shows typically the gap position of the character specified based on distribution of the number w2. In the 1st Embodiment of this invention, it is a figure which shows typically the several image block BL obtained by dividing | segmenting the image of the character area L1. It is a figure which shows typically the division | segmentation table which a tablet terminal hold | maintains. It is a table | surface which shows typically the number table which a tablet terminal produces in the 1st Embodiment of this invention. In the 1st Embodiment of this invention, it is a figure which shows typically the 1st number attached | subjected to each of several image block BL. The character strings included in each of the plurality of image blocks BL shown in FIG. 10 are represented. In the 1st Embodiment of this invention, it is a table | surface which shows typically the number sequence table regarding several image block BL. In the 1st Embodiment of this invention, it is a figure which shows typically each of several image block BL after changing the arrangement | sequence order according to a 2nd number. In the 1st Embodiment of this invention, it is a figure which shows typically the state which inserted the image for connection between each of several image block BL after changing the arrangement | sequence order according to a 2nd number. In the 1st Embodiment of this invention, it is a figure which shows typically an example of the encryption image produced based on the image of the character area L1. In the 1st Embodiment of this invention, it is a figure which shows typically the other example of the encryption image produced based on the image of the character area L1. In the 1st Embodiment of this invention, it is a figure which shows typically the other example of the encryption image produced based on the image of the character area L1. In the 1st Embodiment of this invention, it is a figure which shows typically the data after OCR which the OCR terminal produced. In the 1st Embodiment of this invention, it is a figure which shows typically the some character string obtained by dividing | segmenting the data after OCR. It is a figure which shows typically the text data in the character area L1 produced in the 1st Embodiment of this invention. It is a figure which shows typically the attachment method of the text data in the 1st Embodiment of this invention. It is a flowchart which shows operation | movement of the character image processing system in the 1st Embodiment of this invention. It is a sequence diagram which shows the outline | summary of operation | movement of the character image processing system in the 2nd Embodiment of this invention. It is a figure which shows typically the structure of the encryption matrix in the 2nd Embodiment of this invention. FIG. 25 is a number table showing the relationship between the first number and the second number shown in the encryption matrix of FIG. It is a figure which shows typically the image of the character area L1 in the 2nd Embodiment of this invention. In the 1st Embodiment of this invention, it is a figure which shows typically the several image block BL obtained by dividing | segmenting the image of the character area L1. It is a figure which shows typically the encryption image produced in the 2nd Embodiment of this invention. It is a figure which shows typically the data after OCR produced | generated in the 2nd Embodiment of this invention. In the 2nd Embodiment of this invention, the screen of the binary editor which displayed the character contained in the data after OCR is shown typically. It is a flowchart which shows operation | movement of the character image processing system in the 2nd Embodiment of this invention. It is a sequence diagram which shows the outline | summary of operation | movement of the character image processing system in the modification of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  In the following embodiment, a case where the information processing apparatus (first information processing unit) is configured by a tablet terminal will be described. The information processing apparatus may be an MFP, a PC (Personal Computer), a mobile phone, a facsimile machine, a printer, or a copier. The OCR device only needs to perform OCR processing, and may be, for example, a server, a PC, or a mobile phone.

  [First Embodiment]

    (Configuration of document image processing system)

  FIG. 1 is a block diagram conceptually showing the structure of the document image processing system in the first embodiment of the present invention.

  Referring to FIG. 1, document image processing system in the present embodiment includes MFP 100 and tablet terminal 200 (an example of a first information processing unit), OCR terminals 300-1 and 300-2 (second information processing). An example). MFP 100 and tablet terminal 200 are connected to each other through an intranet 401 in the office, for example. The intranet 401 is connected to the Internet (external network) 402. Each of MFP 100 and tablet terminal 200 is connected to each of OCR terminals 300-1 and 300-2 through intranet 401 and Internet 402. The tablet terminal 200 can be wirelessly connected to the Internet 402 through a repeater (not shown), for example, when taken out of the office by a user.

  The intranet 401 uses a dedicated line such as a wired or wireless LAN. The intranet 401 connects various devices using a TCP / IP (Transmission Control Protocol / Internet Protocol) protocol. Devices connected to the intranet 401 can communicate with each other.

  The Internet 402 uses a WAN (Wide Area Network). Devices connected to the Internet 402 can communicate with each other. Furthermore, a device connected to the intranet 401 can communicate with a device connected to the Internet 402.

  Each of the OCR terminals 300-1 and 300-2 provides an OCR processing service to the user via the Internet 402. The user receives a service provided by each of the OCR terminals 300-1 and 300-2 through the tablet terminal 200 or the like.

  The MFP 100 includes a CPU 110, a ROM (Read Only Memory) 120, a RAM (Random Access Memory) 130, a storage unit 140, a network I / F 150, an image reading unit 160, a PDF creation unit 170, a character area extraction. A unit 180, an operation panel 190, and an image forming unit 195. CPU 110 is mutually connected to ROM 120, RAM 130, storage unit 140, network I / F 150, image reading unit 160, PDF creation unit 170, character area extraction unit 180, operation panel 190, and image forming unit 195. .

  CPU 110 controls MFP 100 as a whole. The ROM 120 stores a control program executed by the CPU 110. The RAM 130 is a working memory for the CPU 110. The storage unit 140 stores (holds) various types of information. The network I / F 150 communicates with an external device via the intranet 401 or the Internet 402. The image reading unit 160 optically reads a document image. The PDF creation unit 170 creates a PDF file of the image read by the image reading unit 160. The character area extraction unit 180 extracts an image of a character area, which is an area where characters are displayed, from the read image. The operation panel 190 includes a display unit, software keys, hardware keys, and the like. The operation panel 190 displays various information and accepts various operations.

  The image forming unit 195 executes a print job. The image forming unit 195 is roughly composed of a toner image forming unit, a fixing device, a paper transport unit, and the like. The image forming unit 195 forms (prints) an image on a sheet by, for example, electrophotography. The image forming unit 195 is configured to be able to form a color image on a sheet by synthesizing four color images by a so-called tandem method. The toner image forming unit includes a photoconductor provided for each color of C (cyan), M (magenta), Y (yellow), and K (black), and an intermediate in which a toner image is transferred (primary transfer) from the photoconductor. The image forming apparatus includes a transfer belt and a transfer unit that transfers an image from the intermediate transfer belt to a sheet (secondary transfer). The fixing device has a heating roller and a pressure roller. The fixing device conveys the sheet on which the toner image is formed between the heating roller and the pressure roller, and heats and presses the sheet. As a result, the fixing device melts the toner adhering to the paper and fixes it on the paper to form an image on the paper. The paper transport unit includes a paper feed roller, a transport roller, and a motor that drives them. The paper transport unit feeds paper from the paper feed cassette and transports the paper inside the housing of MFP 100. The paper transport unit discharges the paper on which the image has been formed from the housing of the MFP 100 to a paper discharge tray or the like.

  The tablet terminal 200 includes a CPU 210, a ROM 220, a RAM 230, a storage unit 240, a network I / F 250, an operation panel 260, an encryption unit 270, a decryption unit 280, and a PDF editing unit 290. Yes. CPU 210 is mutually connected to ROM 220, RAM 230, storage unit 240, network I / F 250, operation panel 260, encryption unit 270, decryption unit 280, and PDF editing unit 290.

  The CPU 210 controls the entire tablet terminal 200. The ROM 220 stores a control program executed by the CPU 210. The RAM 230 is a working memory for the CPU 210. The storage unit 240 stores (holds) various kinds of information such as a software program for creating a searchable PDF and a division table, a number table, or an encryption matrix, which will be described later. The network I / F 250 communicates with an external device via the intranet 401 or the Internet 402. The operation panel 260 displays various information and accepts various operations. The encryption unit 270 encrypts the image of the character area to be transmitted to the OCR terminal 300-1 or 300-2. The decryption unit 280 decrypts the data received from the OCR terminal 300-1 or 300-2 and creates text data. The PDF editing unit 290 adds text data to the image PDF file.

  Each of the OCR terminals 300-1 and 300-2 has an OCR function, and has an OCR site that is a website that provides an OCR processing service. Each of the OCR terminals 300-1 and 300-2 is a different device and has different OCR sites. Each of the OCR terminals 300-1 and 300-2 includes a CPU 310, a ROM 320, a RAM 330, a storage unit 340, a network I / F 350, an OCR processing unit 360, an encryption unit 370, and a decryption unit 380. Is included. CPU 310 is mutually connected to ROM 320, RAM 330, storage unit 340, network I / F 350, OCR processing unit 360, encryption unit 370, and decryption unit 380.

  The CPU 310 controls the entire OCR terminal. The ROM 320 stores a control program executed by the CPU 310. The RAM 330 is a working memory for the CPU 310. The storage unit 340 stores (holds) various information such as an encryption matrix described later. The storage unit 340 has a storage area (personal folder) for the user of the OCR process. A network I / F 350 communicates with an external device via the Internet 402. The OCR processing unit 360 creates text data by performing OCR processing on the character area image received from the tablet terminal 200. The encryption unit 370 encrypts the data obtained by the OCR process. The decryption unit 380 decrypts the data received from the tablet terminal 200 and creates an image of the original character area.

  Note that the number of tablet terminals, MFPs, and OCR terminals included in the character image processing system is arbitrary. The MFP may be an apparatus having an image reading function.

  The MFP 100 and the tablet terminal 200 are connected via an intranet 401. For this reason, information transmitted and received between the MFP 100 and the tablet terminal 200 is difficult to leak. On the other hand, the tablet terminal 200 and each of the OCR terminals 300-1 and 300-2 are connected by the Internet 402. For this reason, information transmitted and received between the tablet terminal 200 and each of the OCR terminals 300-1 and 300-2 is likely to leak.

    (Outline of operation of character image processing system)

  Next, an outline of searchable PDF operation performed by the character image processing system will be described.

  FIG. 2 is a sequence diagram showing an outline of the operation of the character image processing system according to the first embodiment of the present invention.

  Referring to FIG. 2, the user of the tablet terminal issues a searchable PDF creation instruction through the tablet terminal in a state where the document is set in advance on the document table of the MFP. The tablet terminal accepts a searchable PDF creation instruction (process PR0).

  Upon receiving the searchable PDF creation instruction, the tablet terminal instructs the MFP to read the image of the document and transmit the PDF file (notifies the start of processing) (processing PR1).

  Upon receiving an instruction from the tablet terminal, the MFP optically reads an image of the document using a CCD image sensor or the like, and creates read image data digitized by A / D conversion (process PR2). Next, the MFP extracts an image of the character area from the read image data (process PR3). Subsequently, the MFP creates a PDF file of the read image data (process PR4). Next, the MFP transmits the image of the character area, the coordinates of the character area, and the PDF file of the read image data to the tablet terminal (process PR5).

  Upon receiving the character area image, the character area coordinates, and the PDF file of the read image data, the tablet terminal divides the character area image into a plurality of image blocks (process PR6). Each of the plurality of image blocks includes a plurality of characters. Next, the tablet terminal changes (rearranges) the arrangement order of the plurality of image blocks. Next, the tablet terminal connects the plurality of image blocks after the arrangement order is changed using the connection image. As a result, an encrypted image is created (process PR7). An encrypted image is an image of an encrypted character area. Subsequently, the tablet terminal transmits the encrypted image to the OCR terminal (process PR8).

  When the OCR terminal receives the encrypted image from the tablet terminal, the OCR terminal performs OCR processing on the encrypted image to create post-OCR data (process PR9). The post-OCR data is encrypted text data. Subsequently, the OCR terminal transmits the created post-OCR data to the tablet terminal (process PR10).

  When the tablet terminal receives the post-OCR data from the OCR terminal, the tablet terminal divides the received post-OCR data into a plurality of character strings corresponding to a plurality of image blocks. Next, the tablet terminal rearranges the arrangement order of the plurality of character strings into the arrangement order before the change of the plurality of image blocks, and combines the plurality of character strings. Thereby, text data of the image of the character area is created (process PR11). Thereafter, the tablet terminal pastes the obtained text data on the PDF file of the read image data based on the coordinates of the character area (process PR12). As a result, a searchable PDF is created.

  In the present embodiment, information leakage on the Internet 402 is mainly suppressed by the following two methods.

  1. The image of the character area before the OCR processing is divided into a plurality of image blocks including a plurality of characters, and the arrangement order of the plurality of image blocks is changed.

  2. A dummy image (dummy block) is mixed with a plurality of image blocks sent to each of the OCR terminals 300-1 and 300-2 so that dummy information is mixed with the text data after the OCR processing.

    (Instruction for creating searchable PDF)

  Next, a searchable PDF creation instruction (process PR0 in FIG. 2) will be described in detail.

  FIG. 3 is a diagram schematically showing a screen SR displayed on the operation panel of the tablet terminal in the first embodiment of the present invention.

  With reference to FIG. 3, in this Embodiment, operation which a user performs with a tablet terminal becomes a trigger of operation | movement of searchable PDF conversion. The screen SR of the tablet terminal is a screen that accepts a searchable PDF creation instruction and a security level setting. Screen SR includes “searchable PDF creation” key KY1 and “arrow” keys KY2 and KY3.

  Each time the key KY2 is pressed, the tablet terminal lowers the set security level in the order of 4 → 3 → 2 → 1. Each time the key KY3 is pressed, the tablet terminal increases the set security level in the order of 1 → 2 → 3 → 4.

  When the key KY1 is pressed, the tablet terminal starts creating a searchable PDF at the set security level. The tablet terminal instructs the MFP to read a document and transmit a PDF.

    (Text region image extraction method)

  Next, a method for extracting an image of a character area (process PR3 in FIG. 2) will be described in detail.

  FIG. 4 is a diagram schematically showing the character regions L1, L2, and L3 included in the read image data IM in the first embodiment of the present invention.

  Referring to FIG. 4, in extracting an image of a character area, the MFP performs an area determination process on read image data IM. Thereby, the read image data IM is classified into a halftone dot area N1, a photograph area P1, character areas L1, L2, and L3, and other areas Z1. Then, the MFP specifies each image in the character areas L1, L2, and L3 in the read image data IM. The shape of the character area is arbitrary, but here, it is assumed that three character areas L1, L2, and L3 having a rectangular shape are specified in one page of read image data IM.

  The MFP specifies the coordinates of the specified character area. The specified coordinates are the coordinates of the vertices at both ends of the diagonal line of the character area. Specifically, the MFP specifies the coordinates (x1, y1) of the upper left vertex and the coordinates (x11, y11) of the lower right vertex as the coordinates of the character area L1. The MFP specifies the coordinates (x2, y2) of the upper left vertex and the coordinates (x12, y12) of the lower right vertex as the coordinates of the character area L2. The MFP specifies the coordinates (x3, y3) of the upper left vertex and the coordinates (x13, y13) of the lower right vertex as the coordinates of the character area L3. The coordinates of the specified character area may be arbitrary.

  The MFP transmits the images of the character areas L1, L2, and L3 and the coordinates of the character areas L1, L2, and L3 to the tablet terminal. At this time, the coordinates of each character area are preferably stored in the header portion of the image file of the character area. The MFP also transmits a PDF file of the entire read image data IM to the tablet terminal.

  Each image of the character areas L1, L2, and L3 transmitted to the tablet terminal is processed one by one in order in the subsequent processing.

  In the following description, processing relating to the image of the character region L1 will be taken up, but processing for the images of the character regions L2 and L3 is performed in the same manner as the processing for the image of the character region L1.

  Note that the MFP may extract the entire image of the read image data IM as the character region image without extracting the character region image from the image data.

    (Division method of character area image)

  Next, a method for dividing an image of a character area into a plurality of image blocks (processing PR6 in FIG. 2) will be described in detail.

  FIG. 5 is a diagram schematically showing the distribution of white pixels in the x direction and the y direction in the image of the character region L1 in the first embodiment of the present invention. In FIG. 5, the direction of the side extending in the horizontal direction in the image of the rectangular character region L1 is the x direction (an example of the first direction), and the direction of the side extending in the vertical direction in the image of the character region L1. Is the y direction (an example of the second direction). Here, a case where Japanese characters are included in the character area will be described. However, even when characters in a language other than Japanese are included in the character area, division can be performed in the same manner.

  Referring to FIG. 5, in order to prevent the third party from grasping the leaked content even when the image of the character region L1 leaks, the tablet terminal converts the image of the character region L1 into a plurality of image blocks. To divide. The image of the character area L1 is divided so that the number of characters included in one image block falls within the range of the number of characters such that the contents described in the character area L1 cannot be estimated. Further, the image of the character region L1 is divided so that the boundary positions of the plurality of image blocks are not inside the character. This is because when the boundary position is determined inside the character, the character is interrupted in two, and the interrupted character is not recognized correctly in the OCR process.

  Here, it is assumed that the image of the character region L1 is an image in which characters of colors other than white are displayed on a white background. The tablet terminal has a distribution of the number w1 obtained by integrating the white pixels existing in the x direction from the image of the rectangular character region L1, and the distribution along the y direction (hereinafter may be referred to as a distribution of the number w1). To extract. Further, the tablet terminal has a distribution of the number w2 obtained by integrating white pixels existing in the y direction from the image of the rectangular character region L1, and may be described as a distribution along the x direction (hereinafter, distribution of the number w2). ). The tablet terminal creates a plurality of image blocks by dividing the image of the character region L1 at the boundary position determined based on the distribution of each of the numbers w1 and w2.

  The tablet terminal estimates whether the character drawn in the character area L1 is vertical writing or horizontal writing based on the distribution of the numbers w1 and w2. That is, when the maximum value (peak) of the number w1 appears periodically in the distribution of the number w1, the tablet terminal estimates that the character drawn in the character region L1 is horizontal writing. This is because, when the character drawn in the character region L1 is horizontally written, the gap between the lines causes a periodic maximum value in the distribution of the number w1. On the other hand, when the local maximum value of the number w2 varies periodically in the distribution of the number w2, the tablet terminal estimates that the character drawn in the character region L1 is vertical writing. This is because when the character drawn in the character region L1 is vertically written, the gap between the columns causes a periodic maximum value in the distribution of the number w2. If each of the numbers w1 and w2 varies periodically in the distribution of the numbers w1 and w2, the tablet terminal may estimate that the character drawn in the character area L1 is vertical writing. , It may be presumed to be horizontal writing.

  Here, the maximum value of the number w1 appears periodically in the distribution of the number w1. For this reason, the tablet terminal estimates that the character drawn in the character area L1 is horizontal writing. In this case, based on the distribution of the number w1, the tablet terminal specifies the position where the number w1 has a maximum value (peak) as the interline position YP of the image of the character region L1. The tablet terminal divides the image of the character region L1 into a plurality of lines by dividing the image of the character region L1 at the line spacing position YP.

  Next, the tablet terminal extracts the distribution of the number w2 for each of the plurality of divided lines, and specifies the character gap position based on the distribution of the number w2. In the gap position, the number w2 has a maximum value (peak). For this reason, the tablet terminal specifies the position where the number w2 becomes the maximum value as the gap position based on the distribution of the number w2. The gap position is a candidate for the boundary position of the image block.

  FIG. 6 is a diagram schematically showing the gap positions of characters specified based on the distribution of the number w2 in the first embodiment of the present invention. In FIG. 6, the position of the gap between characters is indicated by the tip of a triangle.

  Referring to FIG. 6, characters such as “I”, “Ri”, “Ko”, “Fu”, or “River” are composed of a plurality of lines separated from each other. In these characters, the number w2 has a maximum value at the position inside the character, and the inside of the character is specified as the gap position. When the gap position inside the character becomes the boundary position of the image block, the character is interrupted in two, and the interrupted character cannot be subjected to OCR processing. In addition, when the gap position inside the character becomes the boundary position of the image block, another image block to be connected is found from the segment of the image block, and the security level may be lowered. As the boundary position of the image block, not the gap position inside the character but the gap position between the characters needs to be determined.

  Therefore, the tablet terminal determines, as the boundary position (the position of “divided OK” in FIG. 6), the gap position in which the gap between the adjacent gap positions is equal to or greater than the threshold among the specified gap positions. As the threshold value, an average value or a standard value of intervals between adjacent gap positions can be employed.

  Specifically, the tablet terminal calculates an interval (distance) D1 of each of the gap positions of a plurality of characters. Next, the tablet terminal marks a place where two or more intervals D1 shorter than the threshold value are continuous. The tablet terminal avoids the gap positions that make up the marked location, and determines the boundary position of the image block from among the gap positions that exist in other portions.

  FIG. 7 is a diagram schematically showing a plurality of image blocks BL obtained by dividing the image of the character region L1 in the first embodiment of the present invention.

  Referring to FIG. 7, the tablet terminal divides each of the plurality of rows at the determined boundary position XP. Thereby, a plurality of image blocks BL are obtained. In the present embodiment, since each of the plurality of image blocks BL is divided at a level including characters, when each of the plurality of image blocks BL is subjected to OCR processing, the characters included in the image block BL are correct. It becomes easy to be recognized.

  FIG. 8 is a diagram schematically illustrating a division table held by the tablet terminal.

  Referring to FIG. 8, the division table is a table showing the relationship between the security level and the number of characters to be divided and the number of dummy blocks. The tablet terminal refers to the division table and determines the number of characters to be divided and the number of dummy blocks according to the set security level. The security level is a security level set through the screen shown in FIG. The number of characters to be divided is the size of an image block (the number of characters included in one image block). The dummy block is a block that is inserted when the arrangement order of a plurality of image blocks is changed and is not related to the image of the character area. The dummy block will be described later.

  The division table stipulates that the higher the security level, the smaller the number of characters to be divided and the larger the number of dummy blocks. For example, when the security level is 1, the number of characters to be divided is nine and the number of dummy blocks is zero. When the security level is 3, the number of characters to be divided is 5, and the number of dummy blocks is 1.

    (How to change the order of multiple image blocks and how to connect them)

  Next, a method for changing the arrangement order of a plurality of image blocks and a connection method (processing PR7 in FIG. 2) will be described in detail.

  FIG. 9 is a table schematically showing a number table created by the tablet terminal according to the first embodiment of the present invention.

  Referring to FIG. 9, the number table is a table schematically showing the relationship between the first number and the second number. Here, for convenience of explanation, it is assumed that the image of the character area is divided into nine (9 divisions) image blocks.

  The tablet terminal assigns a first number to each of the plurality of image blocks. The first number indicates the order of each of the plurality of image blocks before the arrangement order is changed. Specifically, the tablet terminal assigns a first number “1”, “2”, “3”, “4”, “5” to each of the plurality of image blocks according to the arrangement order of the plurality of image blocks. , “6”, “7” ”,“ 8 ”,“ 9 ”are assigned first numbers.

  Next, the tablet terminal randomly rearranges the arrangement order of the first numbers. As a rearrangement method, for example, after random numbers 1 to 9 are generated, only numbers that have not been used before are sequentially adopted. When inserting a dummy block, the tablet terminal next inserts a number larger than the number of divisions (here, “10”) at an arbitrary position. This number corresponds to a dummy block. Since the number indicating the dummy block is larger than the number of divisions, it can be easily distinguished from a plurality of image blocks.

  The tablet terminal assigns a second number to each of the plurality of image blocks according to the obtained number sequence. Here, in the tablet terminal, the second number “3” is assigned to the image block whose first number is “1”, and the number “5” is assigned to the image block whose first number is “2”. And the second number “8” is assigned to the image block whose first number is “3”. The second number indicates the arrangement order of the plurality of image blocks after changing the arrangement order (the arrangement order of the plurality of image blocks in the encrypted image transmitted to the OCR terminal).

  Each time the tablet terminal randomly changes the arrangement order of the plurality of image blocks, it may create a new number table, or the number order of the plurality of image blocks may be changed using the number table previously stored. It may be changed. The tablet terminal holds a number table until text data of at least a character area image is created.

  FIG. 10 is a diagram schematically showing a first number assigned to each of a plurality of image blocks BL in the first embodiment of the present invention. FIG. 11 shows a character string included in each of the plurality of image blocks BL shown in FIG. In the following drawings, a black frame may be attached to the outer periphery of the image block for convenience, but no black frame actually exists.

  Referring to FIG. 10, the number of divisions here is 54. The tablet terminal assigns each first number of 1 to 54 to each of the plurality of image blocks BL obtained by dividing the image of the character region L1. The tablet terminal assumes a horizontal writing and assigns a first number to each of the plurality of image blocks BL. When each of the plurality of image blocks BL is arranged according to the first number, an image of the character region L1 is obtained as shown in FIG. 11, and a sentence having the original meaning is obtained.

  In FIG. 11, the image blocks BL indicated as “(22)”, “(23)”, “(24)”, “(36)”, “(53)”, and “(54)” , Both are blank (not including character) blocks. Among these image blocks BL, those described as “(53)” and “(54)” are blanks at the end of the sentence. When the plurality of image blocks BL include blank image blocks, the tablet terminal may insert dummy character images into some or all blank image blocks.

  FIG. 12 is a table schematically showing a number table related to a plurality of image blocks BL in the first embodiment of the present invention. FIG. 13 is a diagram schematically showing each of the plurality of image blocks BL after the arrangement order is changed according to the second number in the first embodiment of the present invention. In FIG. 13, a first number is displayed in each of the plurality of image blocks BL.

  Referring to FIG. 12, a first number and a second number are assigned to each of the plurality of image blocks BL. For example, a second number “26” is assigned to an image block whose first number is “1”, and a number “36” is assigned to an image block whose first number is “2”. The second number “16” is assigned to the image block whose first number is “3”.

  Referring to FIG. 13, after assigning the first and second numbers, the tablet terminal changes the arrangement order of the plurality of image blocks BL according to the second number (at this time, dummy Insert a block). As a result, the image block BL1 having the first number “1” is arranged at the 26th position. The image block BL2 having the first number “2” is arranged at the 36th position. The image block BL3 having the first number “3” is arranged at the 16th position.

  FIG. 14 schematically shows a state in which a connecting image is inserted between each of the plurality of image blocks BL after the arrangement order is changed according to the second number in the first embodiment of the present invention. FIG. In FIG. 14, character strings included in each of the plurality of image blocks BL shown in FIG. 13 are shown.

  Referring to FIG. 14, the tablet terminal creates an encrypted image by connecting each of the plurality of image blocks BL after the arrangement order is changed. When connecting, the tablet terminal inserts a connecting image such as “+” between each of the plurality of image blocks BL after the arrangement order is changed. Any image can be used as the concatenation image, but it is preferably one that is recognized correctly in the OCR processing at the OCR terminal and that the result of character recognition is known. The connection image is typically an image of a symbol that is not a character.

  A dummy character image “####” is included in each of the blank image block BL4 having the first number “53” and the blank image block BL5 having the first number “54”. Has been inserted. As the dummy character image, an arbitrary image can be used, but it is preferable to use an image on which a symbol with a known result of OCR processing is displayed.

  The created encrypted image corresponds to the encrypted image in the character area L1.

    (Encrypted image transmission method)

  Next, the encrypted image transmission method (process PR8 in FIG. 2) will be described in detail.

  FIG. 15 is a diagram schematically showing an example of an encrypted image created based on the image of the character area L1 in the first embodiment of the present invention. FIG. 16 is a diagram schematically showing another example of the encrypted image created based on the image of the character area L1 in the first embodiment of the present invention. FIG. 17 is a diagram schematically showing still another example of the encrypted image created based on the image of the character area L1 in the first embodiment of the present invention.

  Referring to FIG. 15, the encrypted image only needs to be created based on a plurality of image blocks after the arrangement order is changed. In this example, the tablet terminal creates the encrypted image SD1 as one image including a plurality of image blocks into which the connection images are inserted. In this case, the tablet terminal transmits the encrypted image SD1 to one OCR terminal 300-1 (FIG. 1).

  Referring to FIG. 16, in this example, the tablet terminal divides a plurality of image blocks into which images for connection are inserted for each row, thereby creating each of encrypted images SD1 to SD9 in each row. The tablet terminal transmits encrypted images SD1 to SD9 to one OCR terminal 300-1. Thereby, security can be improved.

  Referring to FIG. 17, in this example, the tablet terminal creates each of the encrypted images SD1 to SD9 of each row by dividing a plurality of image blocks into which the images for connection are inserted for each row. The tablet terminal divides the encrypted images SD1 to SD9 into two for transmission to the two OCR terminals. The tablet terminal transmits, for example, the upper (an example of the first part) encrypted images SD1 to SD5 to the OCR terminal 300-1 (FIG. 17A) and sends it to the OCR terminal 300-2 (FIG. 1). In contrast, encrypted images SD6 to SD9 in the lower part (an example of the second part) are transmitted (FIG. 17B). Thereby, security can be further improved. In other words, even if the arrangement order of encrypted images transmitted to one OCR terminal is obtained by a third party and returned to the correct arrangement order, the encrypted images transmitted to the other OCR terminal are not obtained. The character area image is not completely reproduced by a third party.

  When the tablet terminal creates a plurality of encrypted images, each of the plurality of encrypted images includes a file name indicating information specifying the image of the character area and the position of the encrypted image in the image of the character area. Is preferably attached. For example, in the case of the encrypted image on the third line in the image of the second character area L2 (FIG. 4), it is preferable to assign a file name such as “reg02line03.jpeg”.

    (How to create post-OCR data)

  Next, a method for creating post-OCR data (process PR9 in FIG. 2) will be described.

  When the OCR terminal receives the encrypted image, the OCR terminal stores the received encrypted image in a personal folder in the OCR terminal. This personal folder is a folder assigned in advance to the user of the tablet terminal. When the OCR terminal receives a command from the software of the tablet terminal, the OCR terminal starts OCR processing.

  FIG. 18 is a diagram schematically showing post-OCR data created by the OCR terminal in the first embodiment of the present invention.

  Referring to FIG. 18, the OCR terminal creates post-OCR data OD in text format, and stores post-OCR data OD in the same personal folder where the encrypted image is stored. The post-OCR data corresponds to the encrypted text data of the image in the character area L1.

  The post-OCR data OD is preferably given the same file name (excluding the extension) as the file name of the encrypted image. Specifically, when the OCR process is performed on the encrypted image having the file name “reg02line03.jpeg”, the file name “reg02line03.txt” may be assigned to the post-OCR data OD. preferable. Thereby, it is possible to prevent data from being mixed up.

  After the OCR process, the OCR terminal deletes the encrypted image that is the target of the OCR process. The tablet terminal transmits post-OCR data OD to the tablet terminal together with a notification of completion of OCR processing (conversion completion).

  Note that the tablet terminal manages a series of processes such as reception of encrypted images, OCR processing, and transmission of post-OCR data OD so that one OCR terminal does not perform simultaneously. Thereby, it is possible to prevent data from being mixed up.

    (How to create text data for text area images and paste text data)

  Next, a text data image creation method (process PR11 in FIG. 2) and text data paste method (process PR12 in FIG. 2) of the character area image will be described.

  When receiving the post-OCR data, the tablet terminal changes the arrangement order of the post-OCR data OD based on the number table, thereby creating text data in the character area L1. Text data is created using the software of the tablet terminal.

  FIG. 19 is a diagram schematically showing a plurality of character strings obtained by dividing post-OCR data in the first embodiment of the present invention. FIG. 20 is a diagram schematically showing text data in the character area L1 created in the first embodiment of the present invention.

  Referring to FIG. 19, the tablet terminal decomposes the post-OCR data into a plurality of divisional character strings CS based on the concatenation image “+” included in the post-OCR data. Next, the tablet terminal rearranges the arrangement order of the plurality of character strings CS to the original arrangement order based on the number table. According to the number table shown in FIG. 12, the image block whose first number is “1” has moved to the 26th position (has the second number “26”). Therefore, the tablet terminal changes the arrangement order of the 26th character string CS1 sandwiched between the 25th “+” and the 26th “+” to the first. Similarly, the image block whose first number is “2” has moved to the 36th position. Therefore, the tablet terminal changes the arrangement order of the 36th character string CS2 sandwiched between the 35th “+” and the 36th “+” to the second.

  After returning the arrangement order to the original arrangement order, the tablet terminal deletes the linking image and the dummy block as necessary, and calls the character corresponding to the dummy character image (here, “####”) Character). Thereafter, the tablet terminal connects a plurality of character strings to one character string. Thereby, as shown in FIG. 20, the text data TD of the character region L1 is created.

  FIG. 21 is a diagram schematically showing a text data pasting method according to the first embodiment of the present invention.

  Referring to FIG. 21, the PDF file of read image data IM includes a layer LR1 that is a layer including the read image, and a transparent layer LR2 provided on layer LR1. The tablet terminal pastes the obtained text data TD at a position corresponding to the character area L1 in the transparent layer LR2. Similarly, the tablet terminal pastes the text data obtained from the images of the character areas L2 and L3 at positions corresponding to the character areas L2 and L3 in the transparent layer LR2. As a result, a searchable PDF is created.

    (Flowchart showing the operation of the character image processing system)

  FIG. 22 is a flowchart showing the operation of the character image processing system according to the first embodiment of the present invention.

  Referring to FIG. 22, when receiving the security level setting and execution instruction (S1), the CPU of the tablet terminal transmits a scan execution instruction to the MFP (S3).

  When receiving the scan execution instruction, the CPU of the MFP scans the document (S5), and extracts an image of the character area from the read image data (S7). Next, the CPU of the MFP creates a PDF file of the read image (S9), and transmits the image and coordinates of the document area and the PDF file of the read image to the tablet terminal (S11).

  When the CPU of the tablet terminal receives the image of the document area or the like, it divides the image of the character area into a plurality of image blocks (S13), and changes the arrangement order of the plurality of image blocks (S15). Subsequently, the CPU of the tablet terminal inserts a dummy block or a dummy character image into a plurality of image blocks as necessary, and creates an encrypted image by connecting the plurality of image blocks with a connection image (S17). ). Subsequently, the CPU of the tablet terminal transmits the encrypted image to the OCR terminal (S19).

  The CPU of the OCR terminal performs an OCR process on the encrypted image (S21), and transmits the obtained post-OCR data to the tablet terminal (S23).

  The CPU of the tablet terminal divides the post-OCR data into a plurality of character strings, and restores the arrangement order of the plurality of character strings (S25). Next, the CPU of the tablet terminal removes the connection image from the plurality of character strings, and deletes characters corresponding to the dummy block and the dummy character image as necessary, thereby creating text data (S27). Next, the CPU of the tablet terminal creates a searchable PDF by pasting the text data into the PDF file of the read image (S29), and ends the process.

  [Second Embodiment]

    (Outline of operation of character image processing system)

  In this embodiment, first, an outline of the searchable PDF operation performed by the character image processing system will be described.

  FIG. 23 is a sequence diagram showing an outline of the operation of the character image processing system according to the second embodiment of the present invention.

  Referring to FIG. 23, of the operations of the character image processing system according to the present embodiment, from the process in which the tablet terminal accepts a searchable PDF creation instruction (process PR0 in FIG. 2), Up to the process (process PR5 in FIG. 2) to be transmitted to the terminal is the same as the operation (FIG. 2) in the first embodiment. Therefore, the description will not be repeated.

  When the tablet terminal receives the image of the character area, the coordinates of the character area, and the PDF file of the read image data, the tablet terminal divides the image of the character area into a plurality of image blocks. Then, the tablet terminal changes the order of the plurality of image blocks based on the encryption matrix (an example of the first relationship information). As a result, an encrypted image is created (process PR11). Subsequently, the tablet terminal transmits the encrypted image and the encryption matrix to the OCR terminal (process PR12).

  When the OCR terminal receives the encrypted image from the tablet terminal, the OCR terminal restores (restores) the encrypted image to the character area image based on the encryption matrix in accordance with the command from the tablet terminal (process PR13). Next, the OCR terminal creates post-OCR data by performing OCR processing on the character area image in accordance with the command from the tablet terminal (processing PR14). Subsequently, the OCR terminal divides the created post-OCR data into a plurality of data pieces having a predetermined number of bytes according to a command from the tablet terminal, and based on the encryption matrix (an example of second relationship information) The arrangement order of the data pieces is changed (rearranged). As a result, encrypted text data is created (process PR15). Next, the OCR terminal transmits the encrypted text data to the tablet terminal (process PR16).

  When the tablet terminal receives the encrypted text data from the OCR terminal, the tablet terminal returns the arrangement order of the plurality of data pieces in the encrypted text data based on the encryption matrix. Thereby, text data of the image of the character area is created (process PR17). Thereafter, the tablet terminal pastes the obtained text data on the PDF file based on the coordinates of the character area (process PR18). As a result, a searchable PDF is created.

    (Configuration of encryption matrix)

  FIG. 24 is a diagram schematically showing the configuration of the encryption matrix in the second embodiment of the present invention. In FIG. 24, some of the elements in the encryption matrix are enlarged, and the density of the enlarged elements is indicated by numerals.

  Referring to FIG. 24, the tablet terminal holds an encryption matrix. The encryption matrix is a two-dimensional random number table, and the encryption matrix changes the arrangement order to each of the coordinates indicating the arrangement order (first number) of the plurality of image blocks before the arrangement order is changed. It is created by arranging each pixel of density indicating the arrangement order (second number) of a plurality of subsequent image blocks.

  The encryption matrix of FIG. 24 includes 128 (= M, M is a natural number) in the vertical direction and 128 (= N, N is a natural number) in the horizontal direction, for a total of 16384 elements.

  Each element of the encryption matrix has coordinates that increase by one from the left to the right when viewed in one row. Each element of the encryption matrix has a larger coordinate as it is in the lower row. That is, the coordinates of the element in the m-th row and the n-th column (m and n are natural numbers satisfying m ≦ M and n ≦ N) are expressed as “(m−1) × M + n”. The coordinates of each element of the encryption matrix indicate a first number.

  Each element is composed of different concentrations divided into 16384 (= M × N) stages. The element density indicates a second number, and the second number increases as the density decreases. Specifically, the darkest pixel indicates the second number “1”, the second darkest pixel indicates the second number “2”, and the lightest element Indicates the second number “16384”.

  FIG. 25 is a number table showing the relationship between the first number and the second number indicated by the encryption matrix of FIG.

  Referring to FIGS. 24 and 25, the element in the first row and the first column (the element whose first number is “1”) is represented by the “5638” -th darkest density. The element in the first row and the second column (the element whose first number is “2”) is represented by the “1235” -th darkest density. The element in the first row and the third column (the element whose first number is “3”) is represented by the “755” th darkest density. The element in the first row and the fourth column (the element whose first number is “4”) is represented by the “6171” th darkest density.

  The encryption matrix can be created by using a technique for creating a dither matrix of an FM (Frequency Modulation) screen.

  The tablet terminal holds in advance a plurality of encryption matrices having elements such as M × N = 8a × 8a, 8a × 10a, 8a × 12a, and 8a × 14a (a = 1 to about 16). An encryption matrix to be used when changing the arrangement order of the plurality of image blocks may be selected according to the size of the image in the region.

  When the tablet terminal selects an encryption matrix to be used from among a plurality of encryption matrices, the OCR terminal similarly holds a plurality of encryption matrices in advance, and the tablet terminal transmits an encrypted image. The OCR terminal may be notified of information specifying the selected encryption matrix. The tablet terminal may transmit the selected encryption matrix together with the encrypted image to the OCR terminal.

  The tablet terminal may create a new encryption matrix when creating the encrypted image, and transmit the created encryption matrix to the OCR terminal together with the encrypted image.

    (How to create an encrypted image)

  Next, a method for creating an encrypted image (process PR11 in FIG. 23) will be described.

  FIG. 26 is a diagram schematically showing an image of the character area L1 in the second embodiment of the present invention. FIG. 27 is a diagram schematically showing a plurality of image blocks BL obtained by dividing the image of the character region L1 in the first embodiment of the present invention.

  Referring to FIGS. 26 and 27, the tablet terminal divides the image of character region L1 into a plurality of image blocks BL such that each of the plurality of image blocks BL has a size smaller than the characters in the character region. To do. Here, it is assumed that the image of the character region L1 composed of SX × SY pixels is divided into a plurality of image blocks BL having BX (BY is a natural number) × BY (BY is a natural number) pixels. .

  The tablet terminal has an image block size (BX and BY) so that the entire image of the character area L1 is covered with M × N image blocks (the number of elements in the vertical and horizontal directions of the encryption matrix). Value). In other words, the tablet terminal determines the minimum BX and BY that satisfy SX ≦ M × BX and SY ≦ N × BY.

  Note that the image block size (BX × BY) is expressed as BX = x × sf (pieces) and BY = y × sf (pieces). In this notation method, one image block multiplies a rectangle of x × y = 1 × 1, 1 × 2, 1 × 3, or 1 × 4... By a scale factor sf (sf = natural number). It has the shape obtained by this.

  The size of the image block is determined according to the security level. That is, the higher the security level, the smaller the image block size. As the image block size decreases, it becomes more difficult to connect the image blocks based on the design of the image block, and security can be improved.

  In FIG. 27, the image of the character region L1 is composed of 1792 × 896 pixels, and the image of the character region L1 is a plurality of pixels composed of 14 × 7 (= BX × BY) pixels. It is divided into image blocks. In this case, one character is divided into approximately 3 × 6 = 18 image blocks. In this case, it is unlikely that two adjacent character portions are included in one image block. Therefore, it is impossible to estimate the arrangement order of characters from one image block.

  FIG. 28 is a diagram schematically showing an encrypted image created in the second embodiment of the present invention.

  Referring to FIG. 28, the tablet terminal then changes the arrangement order of the plurality of image blocks BL based on the encryption matrix, and combines each of the number of image blocks after the arrangement order is changed. Thereby, the encrypted image SD is created. The encrypted image SD is such that it cannot be determined whether or not characters are included. The tablet terminal transmits the encrypted image SD to the OCR terminal. Note that the tablet terminal may sequentially transmit to the OCR terminal in the changed arrangement order without combining each of the number of image blocks after the arrangement order is changed.

    (How to create encrypted text data)

  Next, a method for creating encrypted text data (process PR15 in FIG. 23) will be described.

  FIG. 29 is a diagram schematically showing post-OCR data generated in the second embodiment of this invention.

  Referring to FIG. 29, the OCR terminal restores the arrangement order of the plurality of image blocks of the received encrypted image based on the encryption matrix, using software preinstalled in the OCR terminal. Thereby, the image of the original character area L1 is restored. Then, the OCR terminal performs OCR processing on the restored image. As a result, post-OCR data OD is created. The created post-OCR data is text data of characters included in the image of the character area L1.

  Note that the OCR terminal may previously hold the same encryption matrix as the encryption matrix held by the tablet terminal. The tablet terminal may transmit the encryption matrix together with the encrypted image to the OCR terminal.

  FIG. 30 schematically shows a binary editor screen displaying characters included in post-OCR data in the second embodiment of the present invention.

  Referring to FIG. 30, next, the OCR terminal divides the post-OCR data into a plurality of data pieces having a predetermined data amount, and changes the arrangement order of the plurality of data pieces based on the encryption matrix.

  Specifically, the OCR terminal displays characters included in the post-OCR data on the screen BS of the binary editor. The screen BS includes a region RG1 and a region RG2. Area RG1 is an area where characters included in post-OCR data are displayed. Area RG2 is an area in which a Shift-JIS format 2-byte binary code corresponding to characters included in post-OCR data is displayed.

  The OCR terminal divides the binary code displayed in the region RG2 into a plurality of data pieces having a predetermined data amount. The data amount of each of the plurality of data pieces is preferably an odd number of bytes (for example, 1 byte or 3 bytes). Thereby, it is avoided that each of the plurality of data pieces is divided in character units, and it is possible to prevent the characters from being decoded from the binary code included in the data pieces.

  In FIG. 30, the characters “[Prior Art ...” included in the post-OCR data correspond to the binary code “81 79 8F 5D 97 88 8B 5A. The character “[” corresponds to the binary code “8179” (where [is a bracket with a corner in the drawing). The letter “subordinate” corresponds to the binary code “8F5D”. The characters “coming” correspond to the binary code “9788”. The character “technique” corresponds to the binary code “8B5A”. The letter “jutsu” corresponds to the binary code “8F70”.

  When dividing into 1-byte data pieces, the OCR terminal divides the binary code displayed in the region RG2 into a plurality of data pieces “81”, “79”, “8F”, “5D”,. Then, the first number is assigned according to the arrangement order of the binary code.

  When dividing into data pieces of 3 bytes, the OCR terminal divides the binary code displayed in the region RG2 into a plurality of data pieces “81798F”, “5D9788”, “8B5A8F”. The first number is assigned according to the arrangement order of

  Next, the OCR terminal changes the arrangement order of the plurality of data pieces based on the encryption matrix. As a result, encrypted text data is created. The OCR terminal transmits the encrypted text data to the tablet terminal.

  The OCR terminal may store a plurality of encryption matrices in advance and select an encryption matrix to be used when changing the arrangement order of the plurality of data pieces according to the total number of characters included in the post-OCR data. .

  When the OCR terminal selects an encryption matrix to be used from among a plurality of encryption matrices, the tablet terminal similarly holds a plurality of encryption matrices in advance, and the OCR terminal transmits the encrypted text data. At this time, information specifying the selected encryption matrix may be notified to the tablet terminal. The OCR terminal may transmit the selected encryption matrix to the tablet terminal together with the encrypted text data.

  The OCR terminal may create a new encryption matrix when creating encrypted text data, and send the created encryption matrix to the tablet terminal together with the encrypted text data.

  Further, the OCR terminal may use the same encryption matrix used when restoring the encrypted image to the image of the character area L1 as the encryption matrix used when changing the arrangement order of the plurality of data pieces. A different one may be used.

  The configuration of the character image processing system in the present embodiment and operations other than those described above are the same as the configuration and operation of the character image processing system in the first embodiment, and therefore description thereof will not be repeated.

    (Flowchart showing the operation of the character image processing system)

  FIG. 31 is a flowchart showing the operation of the character image processing system according to the second embodiment of the present invention.

  Referring to FIG. 31, the character image processing system first performs steps S1 to S13 in the flowchart shown in FIG.

  Following the processing in step S13, the CPU of the tablet terminal creates an encrypted image by changing the arrangement order of the plurality of image blocks based on the encryption matrix (S101). Next, the CPU of the tablet terminal transmits the encrypted image to the OCR terminal (S103).

  Upon receiving the encrypted image, the CPU of the OCR terminal restores the original character area image by returning the arrangement order of the plurality of image blocks based on the encryption matrix (S105). Next, the CPU of the OCR terminal executes OCR processing on the image of the character area (S107), and divides the obtained post-OCR data into a plurality of data pieces (S109). Subsequently, the CPU of the OCR terminal creates encrypted text data by changing the arrangement order of the plurality of data pieces based on the encryption matrix (S111). Next, the CPU of the OCR terminal transmits the encrypted text data to the tablet terminal (S113).

  Upon receiving the encrypted text data, the tablet terminal restores (creates) the text data by returning the arrangement order of the plurality of data pieces based on the encryption matrix (S115). Next, the CPU of the tablet terminal creates a searchable PDF by pasting the text data into the PDF file of the read image (S117), and ends the process.

  [Effect of the embodiment]

  According to the above-described embodiment, when performing OCR processing using an OCR site, the tablet terminal transmits an encrypted image of a character area image to be subjected to OCR processing to the OCR site. Send. The OCR terminal transmits the encrypted text data of the character area image to the tablet terminal. Thereby, leakage of confidential information can be prevented, and the security of OCR processing at an external OCR site can be improved.

  In addition, when creating a searchable PDF using an OCR site, a searchable PDF can be created by pasting text data based on the coordinates of a character area in image data.

  In particular, according to the first embodiment, a plurality of image blocks are created by dividing image data at a gap position between characters, and encryption is performed by changing the arrangement order of the created image blocks. An image is created. The text data obtained by the OCR process is encrypted and transmitted to the tablet terminal. As a result, even if the encrypted image is illegally obtained by a third party, it is difficult for the third party to grasp the character string included in the original image data. In addition, since each of the plurality of image blocks includes characters, the accuracy of the OCR process is improved.

  In particular, according to the second embodiment, by dividing the image data into a plurality of image blocks having a size smaller than the characters in the image data, a plurality of image blocks are created. By changing the arrangement order, an encrypted image is created. The text data obtained by the OCR process is encrypted and transmitted to the tablet terminal. Thereby, even if the encrypted image is illegally obtained by a third party, it becomes difficult for the third party to grasp even the characters included in the image data.

  [Others]

  FIG. 32 is a sequence diagram showing an outline of the operation of the character image processing system in the modification of the present invention.

  Referring to FIG. 32, in this modification, the MFP performs each process performed by the tablet terminal in the first embodiment described above. The character image processing system does not include a tablet terminal, but includes only an MFP (an example of an information processing apparatus) and an OCR terminal. The operation of the character image processing system in this modification will be described below.

  The user of the MFP issues a searchable PDF creation instruction through the operation panel of the MFP in a state where the document is set in advance on the document table of the MFP. The MFP accepts a searchable PDF creation instruction (process PR0).

  When the MFP receives a searchable PDF creation instruction, the MFP optically reads an image of a document and creates read image data (process PR2). Next, the MFP extracts a character area image from the read image data (process PR3). Subsequently, the MFP creates a PDF file of the read image data (process PR4). Next, the MFP divides the image of the character area into a plurality of image blocks (process PR6). Next, the MFP changes the arrangement order of the plurality of image blocks. Next, the MFP connects the plurality of image blocks after the arrangement order is changed using a connection symbol to create an encrypted image (process PR7). Subsequently, the MFP transmits the encrypted image to the OCR terminal (process PR8).

  When the OCR terminal receives the encrypted image from the MFP, the OCR terminal performs OCR processing on the encrypted image to create post-OCR data (processing PR9). Subsequently, the OCR terminal transmits the created post-OCR data to the MFP (process PR10).

  When the MFP receives the post-OCR data from the OCR terminal, the MFP divides the received post-OCR data into a plurality of character strings in units of image blocks. Next, the MFP rearranges the arrangement order of the plurality of character strings into the arrangement order before the change of the plurality of image blocks, and combines the plurality of character strings. Thereby, text data of the image of the character area is created (process PR11). Thereafter, the MFP pastes the obtained text data on the PDF file based on the coordinates of the character area (process PR12). As a result, a searchable PDF is created.

  Similarly, the tablet terminal may be omitted by the MFP performing each operation performed by the tablet terminal in the second embodiment.

  The image data to be processed by the present invention is not limited to the PDF format. The present invention can be applied when OCR processing is performed on image data of any format. The image data to be subjected to the OCR process may be image data held by the information processing apparatus in addition to the read image data obtained by reading a document.

  The above-described embodiments can be combined with each other. For example, in the first embodiment, the post-OCR data obtained by OCR processing is divided into a plurality of data pieces by the same method as in the second embodiment, using the same method as in the second embodiment. And you may transmit to a tablet terminal. In the first embodiment, the relationship between the first number and the second number may be recorded using the encryption matrix as in the second embodiment, or the second embodiment In the embodiment, the relationship between the first number and the second number may be recorded using a number table as in the first embodiment.

  The processing in the above-described embodiment may be performed by software or may be performed using a hardware circuit. It is also possible to provide a program for executing the processing in the above-described embodiment, and record the program on a recording medium such as a CD-ROM, a flexible disk, a hard disk, a ROM, a RAM, or a memory card and provide it to the user. You may decide to do it. The program is executed by a computer such as a CPU. The program may be downloaded to the apparatus via a communication line such as the Internet.

  The above-described embodiment is to be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

100 MFP (Multifunction Peripheral)
110, 210, 310 CPU (Central Processing Unit)
120, 220, 320 ROM (Read Only Memory)
130, 230, 330 RAM (Random Access Memory)
140, 240, 340 Storage unit 150, 250, 350 Network I / F
160 Image Reading Unit 170 PDF (Portable Document Format) Creation Unit 180 Character Area Extraction Unit 190 Operation Panel 195 Image Formation Unit 200 Tablet Terminal 260 Operation Panel 270, 370 Encryption Unit 280, 380 Encryption Decryption Unit 290 PDF Editing Unit 300-1 , 300-2 Optical Character Recognition (OCR) terminal 360 OCR processing unit 401 Intranet 402 Internet BL, BL1, BL2, BL3, BL4, BL5 Image block BS Binary editor screen CS, CS1, CS2 Character string D1 Multiple character gaps Distance between each position (distance)
IM scanned image data KY1, KY2, KY3 key L1, L2, L3 Character area LR1 layer LR2 Transparent layer N1 Halftone dot area OD OCR data P1 Photo area RG1, RG2 Binary editor screen area SD, SD1, SD2, SD3 , SD4, SD5, SD6, SD7, SD8, SD9 Encrypted image SR Screen displayed on tablet operation panel TD Text data XP Boundary position YP Line spacing Z1 Other area

Claims (18)

A character image processing system comprising a first information processing unit and a second information processing unit having an OCR (Optical Character Recognition) function capable of communicating with the first information processing unit via a network,
The first information processing unit includes:
Image block creating means for dividing a character area in the image data into a plurality of image blocks;
Arrangement order changing means for changing the arrangement order of the plurality of image blocks;
Inserting means for inserting a connecting image between each of the plurality of image blocks after the arrangement order is changed by the arrangement order changing means;
An encrypted image created on the basis of the plurality of image blocks after the arrangement order is changed by the arrangement order changing means, the encrypted image including the plurality of image blocks into which the connection images are inserted And a first transmission means for transmitting to the second information processing unit,
The second information processing unit
OCR processing means for creating first text data by performing OCR processing on the encrypted image;
Second transmitting means for transmitting post-OCR data including the first text data to the first information processing unit;
The first information processing unit further includes:
Based on the post-OCR data, the post-OCR data is decomposed into a plurality of character strings, and characters corresponding to the concatenation image are deleted from the post- OCR data. Creating means for creating text data;
A character image processing system including pasting means for pasting the second text data at a position corresponding to each of character regions in the image data. The first information processing unit further includes character area specifying means for specifying a character area in the image data and specifying coordinates of the character area in the image data,
The character image processing system according to claim 1, wherein the pasting unit pastes the second text data based on the coordinates.   The character image processing system according to claim 2, wherein the pasting unit pastes the second text data at a position corresponding to the character area in the transparent layer of the image data. The first information processing unit further includes level receiving means for receiving a security level setting,
The character according to claim 1, wherein the image block creating unit divides the character region into the plurality of image blocks having a size determined according to the level received by the level receiving unit. Image processing system. The second information processing unit includes a first OCR device and a second OCR device different from the first OCR device,
The first transmission means transmits a first portion of the encrypted image to the first OCR device, and a second portion of the encrypted image that is different from the first portion is the first portion. The character image processing system according to claim 1, wherein the character image processing system transmits the data to two OCR devices.   The character image processing system according to claim 1, wherein the first information processing unit further includes an image reading unit that creates the image data by reading an image of a document. The image block creating means
A distribution of the number of white pixels accumulated in a first direction which is the direction of one side in the rectangular character region, and a distribution along a second direction perpendicular to the first direction First distribution extracting means for extracting
A distribution of the number of white pixels existing in the second direction in the rectangular character region, the second distribution extracting unit extracting a distribution along the first direction;
Dividing means for creating the plurality of image blocks by dividing a character area in the image data at a position determined based on the distribution extracted by each of the first and second distribution extracting means; The character image processing system according to claim 1, comprising: The dividing means includes
Based on the distribution extracted by the first distribution extracting means, a line spacing specifying means for specifying a line spacing;
Line dividing means for dividing the character area into a plurality of lines by dividing the character area between the lines specified by the line-interval specifying means,
The second distribution extraction unit is a distribution of the number of white pixels existing in the second direction for each of the plurality of rows divided by the row division unit, the first direction To extract the distribution along
The dividing means includes
Based on the distribution extracted by the second distribution extracting means, a gap specifying means for specifying the gap position of the character;
Boundary determining means for determining a boundary position based on the gap position specified by the gap specifying means;
The character image processing system according to claim 7, further comprising: a column direction dividing unit that divides each of the plurality of rows at a boundary position determined by the boundary determining unit.   9. The character according to claim 8, wherein the boundary determining unit determines, as a boundary position, a gap position in which the distance from another adjacent gap position is equal to or greater than a threshold among the gap positions specified by the gap specifying unit. Image processing system. The character image processing system according to claim 1, wherein the connection image is an image that has a known character recognition result and is held in advance by the first information processing unit . The character image processing system according to claim 10 , wherein the connection image is an image of a symbol that is not a character. The first information processing unit includes the plurality of image blocks before the arrangement order is changed by the arrangement order changing unit, and the plurality of image blocks after the arrangement order is changed by the arrangement order changing unit. further comprising character image processing system according to any one of claims 1 to 11, sequence information holding means for holding the relation information showing the relation between each of the sequence of image blocks. The network is the Internet, the character image processing system according to any one of claims 1 to 12. The first information processing unit, a document containing the optically readable image forming apparatus, the character image processing system according to any one of claims 1 to 13. The first information processing unit further includes a terminal separate from the image forming apparatus,
The character image processing system according to claim 14 , wherein the first transmission unit transmits the encrypted image from the terminal to the second information processing unit. The first information processing unit generates the encrypted image based on image data optically read, the character image processing system according to any one of claims 1 to 15. An information processing apparatus that communicates with an OCR (Optical Character Recognition) apparatus,
Image block creating means for dividing a character area in the image data into a plurality of image blocks;
Arrangement order changing means for changing the arrangement order of the plurality of image blocks;
Inserting means for inserting a connecting image between each of the plurality of image blocks after the arrangement order is changed by the arrangement order changing means;
An encrypted image created on the basis of the plurality of image blocks after the arrangement order is changed by the arrangement order changing means, the encrypted image including the plurality of image blocks into which the connection images are inserted Transmitting means for transmitting to the OCR device;
Receiving means for receiving post-OCR data including first text data created by performing OCR processing based on the encrypted image from the OCR device;
Based on the post-OCR data, the post-OCR data is decomposed into a plurality of character strings, and characters corresponding to the concatenation image are deleted from the post- OCR data. Creating means for creating text data;
An information processing apparatus comprising: pasting means for pasting the second text data at positions corresponding to character regions in the image data. A control program for an information processing device that communicates with an OCR (Optical Character Recognition) device,
An image block creating step for dividing a character area in the image data into a plurality of image blocks;
An arrangement order changing step for changing the arrangement order of the plurality of image blocks;
An insertion step of inserting a connecting image between each of the plurality of image blocks after changing the arrangement order in the arrangement order changing step;
An encrypted image created on the basis of the plurality of image blocks after the arrangement order is changed in the arrangement order changing step, the encrypted image including the plurality of image blocks into which the connection images are inserted Transmitting to the OCR device;
A receiving step of receiving post-OCR data including first text data created by performing OCR processing based on the encrypted image from the OCR device;
Based on the post-OCR data, the post-OCR data is decomposed into a plurality of character strings, and characters corresponding to the concatenation image are deleted from the post- OCR data. A creation step to create text data;
A control program for an information processing apparatus for causing a computer to execute a pasting step of pasting the second text data at a position corresponding to each of character regions in the image data.