JP2012093895A - Image processing system, image processing method and image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve retrievability and editing performance by superimposing character information generated by recognizing characters contained in an image on an original image when rendering an electronic document retrievable and editable.SOLUTION: An image processing system 1 generating character information by recognizing characters displayed as an image comprises: an information acquisition section 141 for acquiring the image in which the characters are displayed; an OCR controller 142 for generating a plurality of character recognition results by causing a plurality of different OCR processing sections to recognize the characters displayed in the image and generate the character information; an OCR result priority order determination section 147 for determining a priority order of the plurality of character recognition results based on difference between each of the plurality of character recognition results; and a document generation section for superimposing character strings generated as the plurality of character recognition results on the image as a transparent layer according to the priority order thereof.

Description

  The present invention relates to an image processing apparatus, an image processing method, and an image processing program, and more particularly to improving the usability of an electronic document generated by character recognition.

  In recent years, there has been a tendency to digitize information, and image processing apparatuses such as printers and facsimiles used for outputting digitized information and scanners used for digitizing documents have become indispensable devices. Such an image processing apparatus is often configured as an MFP (Multi Function Peripheral) that can be used as a printer, a facsimile, a scanner, and a copier by providing an imaging function, an image forming function, a communication function, and the like. .

  Among such image processing apparatuses, in a scanner used for digitizing documents, characters included in an image generated by reading are recognized by OCR (Optical Character Recognition) to generate character information, thereby generating an image. As a result, it is possible to search for information in an electronic document and to edit such as copy and paste and correction.

  Such an electronic document includes image information generated by reading a document and character information generated by the OCR process. The character information generated by the OCR process is arranged at a position corresponding to the coordinates where the character is recognized on the image. Furthermore, the character information arranged on the image is added as a transparent layer so as not to be directly recognized by the user.

  When an electronic document generated in such a manner is opened with a corresponding viewer or the like and a character string is searched for, it is displayed as if the characters included in the image are extracted by the search. When the user performs a mouse operation or the like so as to select a character string on the image, the character information added as a transparent layer is selected, so that the character included in the image is selected visually. The character information can be copied.

  However, since it is difficult to achieve a recognition rate of 100% in OCR, a character string included in an image is not recognized as character information in the electronic document and may not be extracted by a search. obtain. In addition, as a result of erroneously recognizing characters, a character string copied by selection may be different from a character string displayed as an image.

  There are many types of issues like Japanese, and there are many characters with similar shapes, such as Katakana “Ro” and Kanji “Mouth”, Katakana “E” and Kanji “Kou”. This is particularly noticeable in languages that do.

  In addition, the accuracy of character recognition by OCR depends on the compatibility of the OCR algorithm with character types such as font type and font size on the document, and language types such as Japanese and English. For example, some methods of OCR algorithm have high alphabet recognition accuracy, but kana and kanji recognition accuracy is low. Some methods of OCR algorithm have high kana and kanji recognition accuracy and low alphabet recognition accuracy. There can be.

  Therefore, it is preferable to improve the recognition accuracy by selecting an OCR algorithm according to the character type or language type. However, in the case of a document created with a plurality of language types or character types, OCR is performed for each page or each word. It was difficult to choose an algorithm.

  In order to improve the accuracy of character recognition by such OCR, a method has been proposed in which OCR processing is performed using a plurality of OCR algorithms, and one of the most probable OCR recognition results is selected as a recognition result. (For example, refer to Patent Document 1). In the technique disclosed in Patent Document 1, one of the most probable OCR recognition results is selected from a plurality of OCR recognition results by a majority decision based on a plurality of OCR recognition results or by complementing characters by language processing.

  In the invention disclosed in Patent Document 1, when selecting one result from a plurality of OCR recognition results, it is not realistic for the user to manually select each time the document is read. Although automatic selection is adopted by majority vote or language processing, there is a difficulty in selecting a correct recognition result. Therefore, an incorrect recognition result may be selected, and in that case, the problem is still not solved.

  In addition, in order to improve the accuracy when selecting a correct recognition result from among a plurality of recognition results, it is necessary to execute advanced processing, which increases processing load and processing time. Moreover, the problem that it is difficult to discriminate characters based on the shape of the character in a language where there are many characters with similar shapes as described above cannot be solved.

  The present invention has been made to solve such problems, and by superimposing character information generated by recognizing characters included in an image on the original image, searchability and editing can be performed on an electronic document. The purpose is to improve searchability and editability when the possibility is given.

  In order to solve the above-described problem, an aspect of the present invention is an image processing apparatus that recognizes characters displayed as an image and generates character information, and an image acquisition unit that acquires an image on which characters are displayed A character recognition unit that generates a plurality of character recognition results by recognizing characters displayed in the acquired image by a plurality of different character recognition methods and generating character information; and the plurality of character recognitions Based on the difference between the results, the priority determining unit that determines the priority of the plurality of character recognition results, and the character strings generated as the plurality of character recognition results are acquired according to the determined priority. And an image superimposing unit that superimposes as a transparent layer on the image.

  According to another aspect of the present invention, there is provided an image processing method for recognizing a character displayed as an image and generating character information, acquiring the image on which the character is displayed, storing the image in a storage medium, A plurality of character recognition results are generated by recognizing characters displayed in the acquired image by different character recognition methods, generating character information, and storing the information in a storage medium. Based on each difference, the priority order of the plurality of character recognition results is determined and stored in a storage medium, and the character strings generated as the plurality of character recognition results are acquired according to the determined priority order. It is characterized by being superimposed on the image as a transparent layer and stored in a storage medium.

  According to still another aspect of the present invention, there is provided an image processing program for recognizing a character displayed as an image to generate character information, and acquiring the image on which the character is displayed and storing the image in a storage medium Generating a plurality of character recognition results by recognizing characters displayed in the acquired image by a plurality of different character recognition methods and generating character information and storing the character information in a storage medium; and Determining a priority order of the plurality of character recognition results based on a difference between the plurality of character recognition results and storing the priority in the storage medium; and determining the character strings generated as the plurality of character recognition results. And causing the information processing apparatus to execute a step of superimposing a transparent layer on the acquired image according to a priority order and storing the layer on a storage medium.

  According to the present invention, searchability and editability can be obtained when the electronic document is made searchable and editable by superimposing character information generated by recognizing characters included in the image on the original image. Can be improved.

1 is a block diagram schematically illustrating a hardware configuration of an image forming apparatus according to an embodiment of the present invention. 1 is a block diagram illustrating a functional configuration of an image forming apparatus according to an embodiment of the present invention. It is a block diagram which shows the function structure of the image process part which concerns on embodiment of this invention. It is a flowchart which shows operation | movement of the image process part which concerns on embodiment of this invention. It is a figure which shows the input screen of the OCR parameter which concerns on embodiment of this invention. It is a flowchart which shows operation | movement of the image process part which concerns on embodiment of this invention. It is a figure which shows the adjustment aspect of the priority which concerns on embodiment of this invention. It is a figure which shows the superimposition position of the character string which concerns on embodiment of this invention. It is a figure which shows the superimposition aspect of the transparent layer of the character string which concerns on embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the present embodiment, an image processing apparatus that recognizes characters included in an image of a document read by scanning and generates character information and superimposes the character information on the image as a transparent layer to generate an electronic document is taken as an example. explain. The image processing apparatus according to the present embodiment performs character recognition by using a plurality of character recognition methods, generates a plurality of character recognition results, and determines the priority order of each character recognition result based on the contents of the character recognition results. The function of superimposing all character recognition results in the order of the determined priorities is included as a characteristic function.

  FIG. 1 is a block diagram illustrating a hardware configuration of the image processing apparatus according to the present embodiment. The image processing apparatus 1 according to the present embodiment is a multifunction machine that can be used as a printer, a facsimile machine, a scanner, and a copying machine by including an imaging function, an image forming function, a communication function, and the like. The image processing apparatus 1 includes an engine for realizing a scanner, a printer, and the like in addition to the hardware configuration shown in FIG.

  As shown in FIG. 1, the image processing apparatus 1 according to the present embodiment includes the same configuration as a general server, PC, or the like. That is, the image processing apparatus 1 according to this embodiment includes a CPU (Central Processing Unit) 10, a RAM (Random Access Memory) 20, a ROM (Read Only Memory) 30, an HDD (Hard Disk Drive) 40, and an I / F 50. 80 is connected. Further, an LCD (Liquid Crystal Display) 60 and an operation unit 70 are connected to the I / F 50.

  The CPU 10 is a calculation unit and controls the operation of the entire image processing apparatus 1. The RAM 20 is a volatile storage medium capable of reading and writing information at high speed, and is used as a work area when the CPU 10 processes information. The ROM 30 is a read-only nonvolatile storage medium and stores a program such as firmware. The HDD 40 is a non-volatile storage medium that can read and write information, and stores an OS (Operating System), various control programs, application programs, and the like.

  The I / F 50 connects and controls the bus 80 and various hardware and networks. The LCD 60 is a visual user interface for the user to check the state of the image processing apparatus 1. The operation unit 70 is a user interface such as a keyboard and a mouse for the user to input information to the image processing apparatus 1.

  In such a hardware configuration, a program stored in a storage medium such as the ROM 30, the HDD 40, or an optical disk (not shown) is read into the RAM 20, and the CPU 10 performs calculations according to these programs, thereby configuring a software control unit. The A functional block that realizes the functions of the image processing apparatus 1 according to the present embodiment is configured by a combination of the software control unit configured as described above and hardware.

  Next, the functional configuration of the image processing apparatus 1 according to the present embodiment will be described with reference to FIG. FIG. 2 is a block diagram illustrating a functional configuration of the image processing apparatus 1 according to the present embodiment. As shown in FIG. 2, the image processing apparatus 1 according to the present embodiment includes a controller 100, an ADF (Auto Document Feeder) 101, a scanner unit 102, a paper discharge tray 103, a display panel 104, and a paper feed table. 105, a print engine 106, a paper discharge tray 107, and a network I / F 108.

  The controller 100 includes a main control unit 111, an engine control unit 112, an input / output control unit 113, an image processing unit 114, and an operation display control unit 115. As shown in FIG. 2, the image processing apparatus 1 according to the present embodiment is configured as a multifunction machine having a scanner unit 102 and a print engine 106. In FIG. 2, the electrical connection is indicated by a solid arrow, and the flow of a sheet or a document bundle is indicated by a broken arrow.

  The display panel 104 is an output interface that visually displays the state of the image processing apparatus 1, and when the user directly operates the image processing apparatus 1 as a touch panel or inputs information to the image processing apparatus 1. It is also an input interface (operation unit). The display panel 104 is realized by the LCD 60 and the operation unit 70 shown in FIG.

  The network I / F 108 is an interface for the image processing apparatus 1 to communicate with a device via a network, and an Ethernet (registered trademark) or a USB interface is used. The network I / F 108 is realized by the I / F 50 shown in FIG.

  The controller 100 is configured by a combination of software and hardware. Specifically, a program stored in a nonvolatile storage medium such as the ROM 30 and the nonvolatile memory as well as the HDD 40 and the optical disk is loaded into a volatile memory (hereinafter referred to as a memory) such as the RAM 20, and the CPU 10 performs an operation according to these programs. The controller 100 is configured by a software control unit configured by performing the above and hardware such as an integrated circuit. The controller 100 functions as a control unit that controls the entire image processing apparatus 1.

  The main control unit 110 plays a role of controlling each unit included in the controller 100, and gives a command to each unit of the controller 100. The engine control unit 120 serves as a driving unit that controls or drives the print engine 106, the scanner unit 102, and the like. The image processing unit 130 generates drawing information based on image information to be printed out under the control of the main control unit 110. The drawing information is information for drawing an image to be formed in the image forming operation by the print engine 106 as an image forming unit.

  The image processing unit 130 processes image data input from the scanner unit 102 to generate image data. The image data is information stored in the storage area of the image processing apparatus 1 as a result of the scanner operation. Furthermore, the image processing unit 130 includes an OCR (Optical Character Recognition) function for recognizing characters included as an image in the generated image data and generating character information as a function according to the gist of the present embodiment. This OCR function will be described in detail later.

  The operation display control unit 140 displays information on the display panel 104 or notifies the main control unit 110 of information input via the display panel 104. The input / output control unit 150 inputs information input via the network I / F 108 to the main control unit 110. The main control unit 110 controls the input / output control unit 150 to access the client terminal 2 via the network I / F 108 and the network.

  When the image processing apparatus 1 operates as a printer, first, the input / output control unit 150 receives a print job via the network I / F 108. The input / output control unit 150 transfers the received print job to the main control unit 110. When receiving the print job, the main control unit 110 controls the image processing unit 130 to generate drawing information based on document information or image information included in the print job.

  When the image processing unit 130 generates the drawing information, the image processing unit 130 stores the generated drawing information in the HDD 40, a non-illustrated nonvolatile memory, or the like. The engine control unit 120 reads out the stored drawing information and inputs it to the print engine 106 to cause the sheet to be conveyed from the sheet feeding table 105 to form an image. As a specific aspect of the print engine 106, an image forming mechanism using an ink jet method, an image forming mechanism using an electrophotographic method, or the like can be used. A document on which image formation has been performed by the print engine 106 is discharged to a discharge tray 107.

  When the image processing apparatus 1 operates as a scanner, the operation display control unit 115 or input / output control is performed according to a user operation on the display panel 104 or a scan execution instruction input from an external device via the network I / F 108. The unit 113 transfers a scan execution signal to the main control unit 111. The main control unit 111 controls the engine control unit 112 based on the received scan execution signal.

  The engine control unit 112 drives the ADF 101 and conveys the document to be imaged set on the ADF 101 to the scanner unit 102. In addition, the engine control unit 112 drives the scanner unit 102 and images a document conveyed from the ADF 101. If no original is set on the ADF 101 and the original is directly set on the scanner unit 102, the scanner unit 102 images the set original under the control of the engine control unit 112. That is, the scanner unit 102 operates as an imaging unit.

  In the imaging operation, an imaging element such as a CCD included in the scanner unit 102 optically scans the document, and imaging information generated based on the optical information is generated. The engine control unit 112 transfers the imaging information generated by the scanner unit 102 to the image processing unit 114. The image processing unit 114 generates image information based on the imaging information received from the engine control unit 112 according to the control of the main control unit 111. Image information generated by the image processing unit 114 is stored in a storage medium attached to the image processing apparatus 1 such as the HDD 40.

  The image information generated by the image processing unit 114 is stored in the HDD 40 or the like as it is according to a user instruction or transmitted to an external device such as the document management server 2 via the input / output control unit 113 and the network I / F 108. The Further, when the image processing apparatus 1 operates as a copying machine, the image processing unit 114 generates drawing information based on imaging information received by the engine control unit 112 from the scanner unit 102 or image information generated by the image processing unit 114. To do. Based on the drawing information, the engine control unit 112 drives the print engine 106 as in the case of the printer operation.

  In such an image processing apparatus 1, as described above, the OCR function by the image processing unit 114 is one of the gist according to the present embodiment. Hereinafter, the OCR function of the image processing unit 114 according to the present embodiment will be described. FIG. 3 is a block diagram illustrating a configuration of functions related to OCR among the functions of the image processing unit 114 according to the present embodiment.

  As shown in FIG. 3, the image processing unit 114 according to the present embodiment includes an information acquisition unit 141, an OCR control unit 142, a first OCR processing unit 143, a second OCR processing unit 144, a third OCR processing unit 145, It includes a four OCR processing unit 146, an OCR result order determination unit 147, a document generation unit 148, and a document output unit 149. As shown in FIG. 3, the image processing unit 114 according to the present embodiment includes a plurality of OCR processing units that perform character recognition according to different algorithms, and superimposes each OCR result on an image. It is.

  The information acquisition unit 141 acquires image information generated by the image processing unit 114 according to the execution of the scan, and information such as parameters input by the user via the display panel 104 when executing the scan with OCR. The OCR control unit 142 controls execution of OCR processing by a plurality of OCD processing units based on parameters input by the user via the display panel 104.

  The first OCR processing unit 143 to the fourth OCR processing unit 146 perform character recognition using different OCR algorithms. The OCR result order determination unit 147 determines the priority order of each character recognition result based on the results of the OCR processing by the first OCR processing unit 143 to the fourth OCR processing unit 146. The processing by the OCR result order determination unit 147 is one of the gist according to the present embodiment.

  Based on the determination result by the OCR result order determination unit 147, the document generation unit 148 generates an electronic document by superimposing a transparent layer of character information generated by OCR on the original image. The document output unit 149 outputs the electronic document generated by the document generation unit 148.

  Next, in the image processing apparatus 1 according to the present embodiment, an operation when scanning with OCR is executed will be described with reference to a flowchart of FIG. As shown in FIG. 4, when a scan with OCR is executed, first, the user inputs a scan execution instruction together with parameters and the like for OCR processing via the display panel 104. Thereby, the information acquisition unit 141 of the image processing unit 114 acquires operation information such as an OCR parameter via the operation display control unit 115 and the main control unit 111 (S401).

  FIG. 5 shows an example of a screen for the user to input parameters for OCR processing. As shown in FIG. 5, on the OCR parameter input screen, the user inputs or selects the number of times of OCR, selects the OCR method to be executed, and inputs the priority order of the selected OCR method. These pieces of information are input to the operation table control unit 115 via the display panel 104 and input to the image processing unit 114 by the main control unit 114.

  When the information acquisition unit 141 acquires operation information, the OCR control unit 142 sets the number of OCR, the method, and the priority based on the information (S402). When the scan is executed in accordance with the scan execution instruction and the image information is input to the image processing unit 114, the information acquisition unit 141 acquires the image information (S403). That is, in S403, the information acquisition unit 141 functions as an image acquisition unit. When the information acquisition unit 141 acquires image information, the OCR control unit 142 sets an OCR parameter for the OCR processing unit corresponding to the selected OCR method (S404), and the OCR target according to the OCR method is set. The image is processed (S405) and input, and OCR processing is executed (S406). Here, the OCR control unit 142 and the first OCR processing unit 143 to the fourth OCR processing unit 146 function as a character recognition unit.

  The OCR control unit 142 repeats the processing of S404 to S406 for the set number of times of OCR (S407 / NO), and when the OCR processing is completed for the set number of times of OCR, that is, for all the specified OCR methods (S407 / YES), the OCR result order determination unit 147 adjusts the priority of the character recognition result by each OCR method based on all the OCR results (S408). That is, the OCR result order determination unit 147 functions as a priority order determination unit.

  Here, the details of the processing of S408 will be described with reference to FIG. As shown in FIG. 6, the OCR result order determination unit 147 acquires one processing result from the OCR processing result generated by each OCR processing unit in the priority order adjustment operation of the OCR result (S601). One word is selected in order from the top of the recognized character string (S602).

  The OCR result order determination unit 147 repeats the processing of S601 and S602 until it acquires words one by one for the OCR results by all methods (S603 / NO), and acquires one word at a time for the OCR results by all methods (S603). / YES), it is determined whether the character strings of the respective words match (S604). If the character strings of all the words match according to the determination in S604 (S604 / YES), one text layer is created for the word (S605).

  On the other hand, when the character strings of all the words do not match (S604 / NO), the OCR result order determination unit 147 adjusts the priority for each different word (S609). The process of S609 will be described with reference to FIGS. As shown in FIG. 7A, the character recognition results of each of the four OCR methods are “frog” (all katakana), “kaji nore” (kanji “ka”, kanji “2”, katakana “ Nore ”),“ Frog ”(all Katakana),“ Rikoru ”(Kanji“ Power ”, Kanji“ Ku ”to Katakana“ Le ”).

  As shown in FIG. 7A, as a result of character recognition, in addition to the priority of the OCR method input on the screen shown in FIG. 5, each OCR method indicates a numerical value indicating the certainty of the result of character recognition. Is included. That is, each OCR processing unit generates information indicating the certainty of the character recognition result together with the character recognition result. As shown in FIG. 7B, the OCR result order determination unit 147 increments the priority for each recognition result. In the example of FIG. 7B, since the character recognition results of priority order 1 and priority order 3 are the same, the OCR result determination unit 147 sets the priority of the character recognition results of priority order 1 and priority order 3 to + To do.

  In addition, since the accuracy of the character recognition result of priority 2 is low, the OCR result order determination unit 147 decrements the priority of the character recognition result of priority 2. In addition to the example of FIG. 7B, for example, when there is one result of different character strings in four character recognition results, a mode in which the priority of the character recognition result is negative is conceivable.

  When the priority of the recognition result is adjusted as shown in FIG. 7B, the OCR result order determination unit 147 has the same character recognition result of the priority order 1 and the priority order 3 as shown in FIG. 7C. Therefore, the character recognition result with the higher priority is adopted. Further, as illustrated in FIG. 7C, the OCR result order determination unit 147 switches the priority order based on the priority adjustment result in S <b> 609.

  When the priority is adjusted by changing the priority in the process of S609, the OCR result order determining unit 147 generates a text layer for each different character string in the character recognition result (S610). In the processing of S610, the text layer generated by the OCR result order determination unit 147 is a text layer as shown in FIG. 7C, but its color is transparent. Thereby, when superimposing a text layer on the image produced | generated by the scan, it can hide from a user in a normal browsing mode.

  When the processing of S605 or S610 is completed, the OCR result order determination unit 147 determines whether or not the position where the character string of each character recognition result is recognized, that is, the coordinates of the image information are the same (S606). When the positions of all the character strings are the same (S606 / YES), the OCR result order determining unit 147 determines the position as the position of the text layer, and as shown in FIG. Then, the text layers are stored in the order of the first layer, the second layer, and the third layer (S607). Note that when the process of S605 is performed, only one text layer is generated, so the text layer is stored only in the first layer.

  On the other hand, when the positions of all the character strings are not the same (S606 / NO), the OCR result order determination unit 147 determines the recognition position of the character string having the highest priority as the position of the text layer, and the result is shown in FIG. As shown, the text layers are stored in the order of the first layer, the second layer, and the third layer according to the priority order of each character string (S611).

  The OCR result order determination unit 147 repeats the process from S601 for all the words included in the character recognition result (S608 / NO), and completes the process from S601 for all the words included in the character recognition result (S608). / YES), the process ends. By such processing, the OCR result order adjustment processing according to the present embodiment is completed.

  In S408 of FIG. 4, when the OCR result order adjustment process is completed, the OCR result order determination unit 147 stores each layer storing the text layers according to the priority in the process of FIG. An electronic document is created by superimposing it on the image (S409). That is, the document generation unit 148 functions as an image superimposing unit. The image processing unit 114 repeats the processes from S403 to S409 for all the originals set as reading targets (S410 / NO), and generates the process when the processes of S403 to S409 are completed for all the originals (S410 / YES). The electronic document is output (S411), and the process ends. By such processing, the operation of the image processing unit 114 in the case where scanning with OCR is executed is completed.

  FIG. 9 shows the layer structure of the electronic document generated by the processing of S409 in FIG. As shown in FIG. 9, the electronic document generated in the present embodiment is configured by superimposing the respective layers generated by the process of FIG. 6 on the original image generated by scanning. As described with reference to FIG. 3, the image processing unit 114 according to the present embodiment includes the first OCR processing unit 143 to the fourth OCR processing unit 146 that perform OCR by different methods, and therefore, a maximum of four Layers can be generated. Therefore, the layers superimposed on the original image are four layers from the first layer to the fourth layer.

  As described above, when the image processing unit 114 according to the present embodiment generates an electronic document by performing OCR in accordance with scanning, the character string in each execution result is different by executing OCR by using a plurality of different methods. In this case, all the different character strings are stored and superimposed on the original image in another layer. As a result, when a plurality of different character strings are recognized, a process for selecting any one of the character strings to be superimposed on the original image becomes unnecessary, and the process can be reduced.

  In addition, when any one character string is selected as a character string to be superimposed on the original image, there may be a case where an erroneously recognized character string is selected instead of a correctly recognized character string. In such a case, even if the user browses the electronic document with a viewing application or the like and searches for the character string, the character string that should be extracted is not extracted. On the other hand, in the case of the present embodiment, the character string recognized for correct confirmation and the character string recognized erroneously all hold different recognition results, so that the above problem can be solved.

  Also, in a language where there are many similar-shaped characters, even if it is difficult to distinguish characters based on the character shape, all the recognition results by more character recognition methods are superimposed on the image. As a result, the probability of obtaining a correct recognition result can be improved.

  On the other hand, as shown in FIG. 9, when different character strings are stored in different layers as recognition results, the user wants to copy and paste the character strings when the accurately recognized character strings are stored in the lower layers. Inconvenience occurs in some cases. That is, in the case of the layer structure as shown in FIG. 9, the character string to be selected and copied is only the top layer in a normal browsing application. Therefore, if a correctly recognized character string is stored in a lower layer, the correctly recognized character string cannot be copied.

  In contrast, in the present embodiment, the OCR result order determination unit 147 adjusts the priority of different character recognition results by the process described in FIG. Since it is stored in the layer, a state in which a misrecognized character string is stored in the first layer and a correctly recognized character string is stored in a lower layer is prevented, and the accurately recognized character string is copied. Can be.

1 Image forming apparatus 10 CPU
20 RAM
30 ROM
40 HDD
50 I / F
60 LCD
70 Operation unit 80 Bus 100 Controller 101 ADF
102 Scanner unit 103 Paper discharge tray 104 Display panel 105 Paper feed table 106 Print engine 107 Paper discharge tray 108 Network I / F
110 Main Control Unit 112 Engine Control Unit 113 Input / Output Control Unit 114 Image Processing Unit 115 Operation Display Control Unit 141 Information Acquisition Unit 142 OCR Control Unit 143 First OCR Processing Unit 144 Second OCR Processing Unit 145 Third OCR Processing Unit 146 First Four OCR processing unit 147 OCR result order determination unit 148 Document generation unit 149 Document output unit

JP 2001-22883 A

Claims (9)

An image processing apparatus that recognizes characters displayed as an image and generates character information,
An image acquisition unit for acquiring an image on which characters are displayed;
A character recognition unit that generates a plurality of character recognition results by recognizing characters displayed in the acquired image by a plurality of different character recognition methods and generating character information;
A priority determining unit that determines the priority of the plurality of character recognition results based on the difference between the plurality of character recognition results;
An image processing apparatus comprising: an image superimposing unit that superimposes the character strings generated as the plurality of character recognition results as a transparent layer on the acquired image according to the determined priority order.   The priority order determination unit acquires a plurality of character strings recognized by the plurality of character recognition methods based on the same recognition target, and determines a priority order for each different character string among the plurality of character strings. The image processing apparatus according to claim 1.   The image processing apparatus according to claim 2, wherein the priority order determination unit increases the priority of a character recognition result having another identical character string among the plurality of character strings.   The image processing apparatus according to claim 2, wherein the priority order determination unit lowers the priority of a character recognition result that has no other identical character string among the plurality of character strings.   The image superimposing unit, at the recognition position on the image of the character string having the highest priority among the recognition positions on the image of each of a plurality of character strings generated as a character recognition result based on the same recognition target, The image processing apparatus according to claim 1, wherein the plurality of character strings are superimposed. An operation information acquisition unit that acquires priority designation information that designates the priority of each of the plurality of different character recognition methods according to an operation by a user;
The priority determination unit determines the priority of the plurality of character recognition results by adjusting the priority specified by the acquired priority specification information according to the difference between the plurality of character recognition results. The image processing apparatus according to claim 1, wherein the image processing apparatus includes: The character recognition unit generates information indicating the certainty of the character recognition result together with the character recognition result,
The image processing apparatus according to claim 1, wherein the priority order determination unit adjusts the priority of the character recognition result based on information indicating the certainty of the character recognition result. An image processing method for recognizing characters displayed as an image and generating character information,
Acquire an image displaying the characters and store it in the storage medium.
Recognizing characters displayed in the acquired image by a plurality of different character recognition methods to generate character information and storing it in a storage medium, thereby generating a plurality of character recognition results,
Based on the difference between each of the plurality of character recognition results, a priority order of the plurality of character recognition results is determined and stored in a storage medium;
An image processing method, wherein the character strings generated as the plurality of character recognition results are superimposed on the acquired image as a transparent layer according to the determined priority order and stored in a storage medium. An image processing program for recognizing characters displayed as an image and generating character information,
Acquiring an image displaying characters and storing the image on a storage medium;
Generating a plurality of character recognition results by recognizing characters displayed in the acquired image by a plurality of different character recognition methods to generate character information and storing the character information in a storage medium;
Determining a priority order of the plurality of character recognition results based on the difference between the plurality of character recognition results, and storing them in a storage medium;
Causing the information processing apparatus to execute a step of superimposing the character strings generated as the plurality of character recognition results as a transparent layer on the acquired image in accordance with the determined priority order and storing them in a storage medium. An image processing program characterized by the above.

Images