JP6484974B2 - Information processing apparatus, information processing system, and program - Google Patents

Description

  The present invention relates to an information processing apparatus, an information processing system, and a program.

  2. Description of the Related Art Conventionally, an OCR (Optical Character Recognition) technique for recognizing characters from a document image read by a scanner or the like and converting them into document data is known. A technique for performing OCR processing on a document including a plurality of languages has also been proposed. For example, Patent Document 1 describes a technique for performing OCR processing on a document including a plurality of languages in different languages for each document area.

  Further, a technique for searching for an OCR result that matches a search condition input by a user from a plurality of OCR results is also known (for example, Patent Document 2).

JP 2006-260115 A JP 2004-213091 A

  Here, a document including a plurality of languages includes, for example, a document in which a part of the document is written in Japanese, another part is written in Korean, and another part is written in English. Some documents contain a mixture of Japanese, Korean, and English. Since Japanese and Korean have similar character shapes, the accuracy of character recognition is reduced in the OCR process especially in the latter document. If the accuracy of character recognition is reduced, the search accuracy for the OCR result is also reduced.

  An object of the present invention is to provide an information processing apparatus, an information processing system, and a program capable of improving the accuracy of character recognition for a document including a plurality of languages and improving the search accuracy for a character recognition result.

An information processing apparatus according to claim 1 of the present invention includes a data acquisition unit configured to acquire document image data obtained by reading a document including a plurality of different character sets as an image, and the character set for each of the plurality of character sets. Document data generating means for recognizing the whole of the document image data and converting the document image data into document data, and generating a number of the document data corresponding to the number of the character sets for one document image data, and the document Document data storage means for storing a plurality of the document data generated by the data generation means, search character set determination means for determining a character set of input characters input as a search condition, and stored in the document data storage means more of the document data, the text corresponding to the character set of the search character set determined by the determination means that said input character For the data, characterized in that it comprises a and a search executing means for executing a search of the input character input as the search condition.

  In the information processing apparatus according to claim 2 of the present invention, in the configuration according to claim 1, when the document includes a first character set and a second character set different from each other, the document data generation means The entire document image data is character-recognized based on the first character set and converted into document data to generate first document data, and the document image data is converted based on the second character set. The whole is recognized and converted into document data to generate second document data.

  In the information processing apparatus according to claim 3 of the present invention, in the configuration according to claim 1, when the document includes a first character set and a second character set different from each other, the document data generation unit , Based on the first character set and the third character set having dissimilar character shapes, the entire document image data is character-recognized and converted into document data to generate first document data, Based on the second character set and the fourth character set having dissimilar character shapes, the entire document image data is recognized and converted into document data to generate second document data.

The information processing apparatus according to claim 4 of the present invention, obtained in arrangement according to claim 1, the document data corresponding to the character set of the inputted input character as the search condition, from the document data storage unit If further comprising, character set of the input character input as the search condition the document data acquisition unit that was the fifth character set, the document data acquiring means, from the document data storage unit, the fifth obtain at least one of the document data including the character set, the search executing means, for each acquisition is the document data, perform a search of the input character.

  An information processing apparatus according to claim 5 of the present invention is the configuration according to claim 1, wherein the plurality of different character sets are a plurality of different languages.

An information processing system according to claim 6 of the present invention is an information processing system including a character conversion device and a search device, wherein the character conversion device reads a document including a plurality of different character sets as an image. Data acquisition means for acquiring image data; and for each of the plurality of character sets, the character set is used to recognize the whole of the document image data and convert it into document data. Document data generation means for generating the number of document data corresponding to the number of sets, document data storage means for storing a plurality of the document data generated by the document data generation means, and the search device for the document data anda transmitting means for transmitting to said search device includes: a receiving means for receiving the document data from the character conversion unit, as a search condition A search character set determination unit that determines a character set of the input character that has been input, and the character of the input character determined by the search character set determination unit among the plurality of document data stored in the document data storage unit for the document data corresponding to the set, the search execution means for executing a search for the inputted input character as the search condition, the search result by the search executing means, the user of the user terminal that inputs the search condition And transmitting means for transmitting to.

A program according to claim 7 of the present invention is a data acquisition means for acquiring document image data obtained by reading a document including a plurality of character sets different from each other as an image, and for each of the plurality of character sets, the document image by the character set. It converts the entire data to the character recognition document data, one of the document the relative image data character sets document data generating means for generating the number of the document data corresponding to the number of the previous SL document data generating means Document data storage means for storing a plurality of document data generated by the above, a search character set determination means for determining a character set of input characters input as a search condition, and a plurality of data stored in the document data storage means wherein among the document data, the document de corresponding to the character set of the search character sets the input character determined by the determining means For the data, a program for the search executing means for executing a search for the inputted input character as the search condition, a computer to function as a. This program may be stored in a computer-readable information storage medium such as a CD-ROM or a DVD-ROM.

  According to the configurations of the first, fifth, sixth, and seventh aspects of the present invention, the accuracy of character recognition for a document including a plurality of languages is improved, and the search accuracy for character recognition results is improved.

  According to the configuration of claim 2 of the present invention, the document image data is converted into each document data corresponding to the language included in the document.

  According to the configuration of claim 3 of the present invention, the document image data is converted into one document data based on two languages whose character shapes are dissimilar to each other for each language included in the document.

  According to the configuration of claim 4 of the present invention, when a plurality of document data is acquired, the input character is searched for each document data.

1 is an overall configuration diagram of an information processing system according to an embodiment. It is a block diagram which shows the hardware constitutions of an OCR server and a search server. It is a functional block diagram of an OCR server. It is a figure which shows an example of the table of a dissimilar language group. It is an example of the selection screen with which a user selects a language. It is an operation | movement flowchart of an OCR execution language determination part. It is a table which shows the information of the registered dissimilar language group. It is an operation | movement flowchart of an OCR execution part. It is a table which shows the information of an OCR result. It is a figure which shows typically the relationship between OCR execution operation and an OCR result. It is a functional block diagram of a search server. It is an operation | movement flowchart of a search server. It is a figure which shows an example of a character code corresponding | compatible table.

  An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is an overall configuration diagram of an information processing system according to the present embodiment. As illustrated in FIG. 1, the information processing system 10 includes an information processing apparatus 100, a multifunction peripheral 400, and a user terminal 500. The information processing apparatus 100 includes an OCR server 200 (character conversion device) and a search server 300 (search device). The OCR server 200 and the search server 300 may be provided as separate devices, or these functions may be included in one device.

  The information processing apparatus 100, the multifunction machine 400, and the user terminal 500 are connected to each other via a communication network such as a LAN (Local Area Network) or the Internet.

  The multi-function device 400 includes a copy function, a FAX function, a scanner function, and the like. In the present embodiment, description will be given focusing on a scanner function for reading a document as an image based on a user operation. The multifunction device 400 optically scans an object (document) such as paper to read the document as an image, saves the read document image data in an internal storage unit, and transmits it to the OCR server 200. Note that the MFP 400 may transmit document image data stored in the storage unit to the OCR server 200 based on an instruction from the OCR server 200. A known configuration can be applied to the MFP 400.

  The user terminal 500 is a terminal device used by a user connected to a communication network. The user terminal 500 performs data communication with the OCR server 200, the search server 300, and the multifunction device 400 via a communication network based on a user operation. Thereby, for example, the user operates the user terminal 500 to instruct the OCR server 200 to execute the OCR process, to instruct the search server 300 to execute the search process on the OCR result, or to search from the search server 300. Get results. In addition, the user operates the user terminal 500 to perform various settings related to the OCR processing or change various settings. The user terminal 500 may incorporate software such as a browser and an e-mail client. The user terminal 500 is, for example, a personal computer, a PDA (Personal Digital Assistant), a portable information terminal such as a smartphone, or the like. The number of user terminals 500 included in the information processing system 10 is not limited.

[Configuration of OCR server]
When the OCR server 200 receives document image data from the multi-function peripheral 400, the OCR server 200 executes OCR processing. FIG. 2 is a block diagram illustrating a hardware configuration of the OCR server 200. The OCR server 200 includes a computer including a CPU 21, a memory 22, a storage unit 23, and a communication unit 24. These hardware elements are connected to each other by a bus so as to be able to exchange data. The communication unit 24 performs data communication with the search server 300, the multi-function device 400, and the user terminal 500 via a communication network. The CPU 21 controls each unit of the OCR server 200 and executes various types of information processing. The memory 22 holds various programs and data. A work area for the CPU 21 is also secured in the memory 22. The storage unit 23 stores various data. The storage unit 23 may be provided outside the OCR server 200.

  FIG. 3 is a functional block diagram of the OCR server 200. As shown in FIG. 3, the OCR server 200 functionally includes a data reception unit 201, an OCR execution language determination unit 202, an OCR execution unit 203, and an OCR result transmission unit 204. These elements are realized by the CPU 21 executing a program stored in the memory 22 (see FIG. 2). This program may be installed in the OCR server 200 from a computer-readable information storage medium such as a CD-ROM, DVD-ROM, or memory card, or may be downloaded via a communication network such as the Internet. The storage unit 23 of the OCR server 200 includes a data storage unit 205, a language information storage unit 206, and an OCR result storage unit 207.

  The data receiving unit 201 (data acquisition unit) receives document image data transmitted from the multi-function peripheral 400. The data receiving unit 201 stores the received document image data in the data storage unit 205. The data storage unit 205 temporarily stores document image data before the execution of the OCR process. The data storage unit 205 is configured by a queue, for example, and performs queuing processing.

  The OCR execution language determination unit 202 determines a language (hereinafter referred to as an OCR execution language) for executing the OCR process. For example, when executing OCR processing in Japanese for Japanese document image data, the OCR execution language is Japanese.

  Here, when a plurality of languages (character sets) are mixed in a document, it is known that the accuracy of character recognition in OCR processing is lowered. When languages that do not have similar character shapes are mixed, the character shapes of the respective languages are recognized as the respective languages, so the probability of recognition is high. However, in particular, when the character shapes of a plurality of languages are similar, the probability of erroneous recognition increases and the character recognition accuracy may further decrease. Note that the characters may include symbols and numbers. Therefore, the OCR server 200 according to the present embodiment executes the OCR process based on a language group (hereinafter referred to as a dissimilar language group) in which a plurality of languages whose character shapes are not similar to each other are combined. For example, Japanese, Korean, and Chinese are similar in character shape to each other, but these are not similar to English. Therefore, “Japanese / English”, “Korean / English”, and “Chinese / English” are set as dissimilar language groups. The dissimilar language group is set in advance by an administrator of the OCR server 200, a user of the user terminal 500, or the like. Information on the set dissimilar language group is stored in the language information storage unit 206. FIG. 4 is a diagram illustrating an example of a table of dissimilar language groups. Identification information (ID) of each dissimilar language group is attached to the table. Other than the above languages, for example, Thai can be set as a dissimilar language group whose character shape is not similar to any of English, Japanese, Korean, and Chinese. In addition, Hindu and Burmese may be set as languages having similar character shapes.

  The OCR execution language determination unit 202 extracts a dissimilar language group corresponding to the language selected by the user from the dissimilar language group stored in the language information storage unit 206, and determines it as an OCR execution language group. FIG. 5 is an example of a selection screen for the user to select a language. The selection screen may be displayed on a display unit of the user terminal 500 connected to the OCR server 200 or may be displayed on an operation screen on the OCR server 200. The user selects (checks) a desired language on the selection screen. Hereinafter, the operation flow of the OCR execution language determination unit 202 will be described with a specific example.

  FIG. 6 is an operation flowchart of the OCR execution language determination unit 202. Here, as an example, the user selects “Japanese”, “Korean”, and “English” on the selection screen of FIG.

  First, the OCR execution language determination unit 202 receives an OCR execution language setting instruction from the user terminal 500 (S101). For example, when the user selects “Japanese”, “Korean”, and “English” on the display screen of the user terminal 500 (see FIG. 5), data (setting instruction) indicating the selected content is transmitted to the OCR server 200. , The OCR execution language determination unit 202 receives. Next, the OCR execution language determination unit 202 acquires one language from the selected languages (S102). Here, for example, “Japanese” is acquired.

  Next, the OCR execution language determination unit 202 acquires a dissimilar language group including the acquired language from the dissimilar language group stored in the language information storage unit 206 (see FIG. 4) (S103). Here, “Japanese / English” (ID: JE) of the dissimilar language group including “Japanese” is acquired. Next, the OCR execution language determination unit 202 registers the acquired dissimilar language group in the OCR execution language group (S104).

  Next, the OCR execution language determination unit 202 determines whether there is another language selected by the user (S105). If there is another language, the process returns to S102 and repeats the above processing. Here, through the above processes, the OCR execution language determination unit 202 acquires “Korean” and acquires “Korean / English” (ID: KE) of the dissimilar language group including “Korean”. The language selected by the user includes “English” (see FIG. 5), but “English” is already included in the dissimilar language group (ID: JE, KE). The group is not acquired. Information on the registered dissimilar language group is stored in the language information storage unit 206. FIG. 7 is a table showing information on registered dissimilar language groups.

  When there is no other language selected by the user (NO in S105), the OCR execution language determination unit 202 determines the registered dissimilar language group (ID: JE, KE) as an OCR execution language group (S106). ).

  As described above, the OCR execution language determination unit 202 determines the OCR execution language (OCR execution language group).

  Returning to FIG. 3, the OCR execution unit 203 (document data generation unit) executes the OCR process using the OCR execution language group determined by the OCR execution language determination unit 202 and generates an OCR result (document data). The OCR result transmission unit 204 transmits the OCR result to the search server 300. Hereinafter, the operation flow of the OCR execution unit 203 will be described with a specific example.

  FIG. 8 is an operation flowchart of the OCR execution unit 203. Here, a case where “Japanese / English” and “Korean / English” shown in FIG. 7 are determined as the OCR execution language group will be described as an example.

  First, the data receiving unit 201 receives document image data from the multifunction machine 400 (S201). The received document image data is referred to as “D1”. Next, the OCR execution unit 203 acquires one dissimilar language group from the OCR execution language group (S202). Here, for example, a dissimilar language group of “Japanese / English” is acquired.

  Next, the OCR execution unit 203 executes OCR processing on the document image data D1 using the acquired dissimilar language group (S203). Specifically, when Japanese and English characters are included in the document image data D1, a portion corresponding to Japanese is converted to Japanese, and a portion corresponding to English is converted to English. Since Japanese and English do not have similar character shapes, it is possible to prevent erroneous conversion in which Japanese is converted into English or English is converted into Japanese. When the document image data D1 includes Japanese and Korean characters, the part corresponding to Japanese is converted to Japanese and the part corresponding to Korean is not converted (results in an error), or It is converted to Japanese with similar character shape. As a result, the Japanese part is not converted into Korean, so at least the Japanese part is reliably converted into Japanese.

  Next, the OCR execution unit 203 associates the OCR result (first OCR result; first document data) with the used dissimilar language group information and stores it in the OCR result storage unit 207 (S204). FIG. 9 is a table showing OCR result information stored in the OCR result storage unit 207. Here, the information is stored as an OCR process ID: t1.

  Next, the OCR execution unit 203 determines whether there are other dissimilar language groups in the OCR execution language group (S205). If there are other dissimilar language groups, the process returns to S202 and repeats the above processing.

  Here, the OCR execution unit 203 acquires a dissimilar language group of “Korean / English”, and executes OCR processing again on the document image data D1 using “Korean / English” (S203). . Specifically, when the document image data D1 includes Korean and English characters, the portion corresponding to Korean is converted to Korean, and the portion corresponding to English is converted to English. Since Korean and English do not have similar character shapes, it is possible to prevent erroneous conversion such as Korean being converted into English or English being converted into Korean. If the document image data D1 includes Korean and Japanese characters, the part corresponding to Korean is converted to Korean and the part corresponding to Japanese is not converted (results in an error), or It is converted to Korean with similar character shape. That is, since the Korean part is not converted into Japanese, at least the Korean part is definitely converted into Korean.

  Next, the OCR execution unit 203 associates the OCR result (second OCR result; second document data) with the used dissimilar language group information and stores it in the OCR result storage unit 207 (S204). Here, the information is stored as the OCR process ID: t2 (see FIG. 9). In FIG. 9, the same JOB_ID: JOB001 is registered in each of the OCR process IDs: t1 and t2, but this is one operation of the OCR execution unit 203 (in the above example, two OCRs are performed). Processing). FIG. 10 schematically shows the relationship between the OCR execution operation and the OCR result. JOB_ID also corresponds to document image data. In the above example, since two dissimilar language groups are registered, OCR processing is performed twice (ID: t1, t2) for one OCR execution. When three are registered, OCR processing is performed three times for one OCR execution.

  As described above, the OCR execution unit 203 executes the OCR process. That is, for example, when Japanese and Korean are mixed in a document, the OCR execution unit 203 first executes OCR processing in Japanese and English on the entire document image data, and then performs the same entire document image data. Execute OCR processing in Korean and English. In the result of the first OCR process, the Japanese part is reliably converted into Japanese. Further, in the result of the second OCR process, the Korean portion is surely converted to Korean. Even if there are characters similar to Japanese and Korean in the document, at least the result of OCR processing in the original language reflects the correct converted character. For this reason, characters that are not accurately converted among the characters in the document and are not included in any OCR result will not occur. In other words, all characters in the document can be accurately converted. Therefore, the accuracy of character recognition for a document including a plurality of languages can be improved.

Search server configuration
The search server 300 performs a character string search on the OCR result generated in the OCR server 200 based on the search condition input by the user. FIG. 2 shows a hardware configuration of the search server 300. The search server 300 includes a computer that includes a CPU 31, a memory 32, a storage unit 33, and a communication unit 34. These hardware elements are connected to each other by a bus so as to be able to exchange data. The communication unit 34 performs data communication with the OCR server 200 and the user terminal 500 via a communication network. The CPU 31 controls each unit of the search server 300 and executes various types of information processing. The memory 32 holds various programs and data. A work area for the CPU 31 is also secured in the memory 32. The storage unit 33 stores various data. The storage unit 33 may be provided outside the search server 300.

  FIG. 11 is a functional block diagram of the search server 300. As shown in FIG. 11, the search server 300 functionally includes an OCR result receiving unit 301, an OCR information management unit 302, a search instruction receiving unit 303, a search language determining unit 304, a search processing unit 305, and a search result. A transmission unit 306 is included. These elements are realized by the CPU 31 executing a program stored in the memory 32 (see FIG. 2). This program may be installed in the search server 300 from a computer-readable information storage medium such as a CD-ROM, DVD-ROM, or memory card, or may be downloaded via a communication network such as the Internet. The storage unit 33 of the search server 300 includes an OCR result storage unit 307, an OCR information storage unit 308, and a character information storage unit 309.

  The OCR result receiving unit 301 receives the OCR result from the OCR server 200. The OCR result reception unit 301 stores the received OCR result in the OCR result storage unit 307. The OCR result storage unit 307 temporarily stores the OCR result before the search process is executed. The OCR result storage unit 307 includes, for example, a queue, and performs queuing processing. Here, for example, OCR results (see FIG. 9) with IDs: t1 to t4 are received.

  The OCR information management unit 302 stores the OCR result received by the OCR result reception unit 301 in the OCR information storage unit 308. The information stored in the OCR information storage unit 308 is the same as the OCR result information stored in the OCR result storage unit 207 (see FIG. 9).

  Hereinafter, the search processing of the search server 300 and elements for executing the search processing will be described with reference to the operation flow of FIG. FIG. 12 is an operation flowchart of the search server 300.

  The search instruction receiving unit 303 receives a search instruction from the user terminal 500 (S301). For example, the user inputs a desired search condition (search keyword) on the user terminal 500. When a search keyword is input, a search instruction including the search keyword is transmitted from the user terminal 500 to the search server 300.

  The search language determination unit 304 determines a search target language based on information (search keyword) included in the received search instruction (S302). Specifically, the search language determination unit 304 acquires the character code of the character of the search keyword, refers to the character code correspondence table, and determines the language corresponding to the acquired character code. The character code correspondence table is registered in advance in the character information storage unit 309. FIG. 13 is a diagram illustrating an example of a character code correspondence table. For example, when the character code of the start character of the search keyword is “3040”, the end character code is “309F”, and the character type is “Hiragana”, the search language determination unit 304 sets the search keyword to “Japanese”. Is determined.

  The search processing unit 305 (search execution means) obtains one OCR result corresponding to the language determined by the search language determination unit 304 from all the OCR results stored in the OCR information storage unit 308 (see FIG. 9). Then, a character string search is performed on the acquired OCR result (S304). For example, if four OCR results of ID: t1 to t4 are stored, the OCR execution language group (dissimilar language group) among them, the OCR result of ID: t1 including “Japanese” is acquired, Perform a string search. If there is another OCR result corresponding to the determined language (YES in S305), the process returns to S303 and the above process is repeated. Here, an OCR result of ID: t3 including “Japanese” in the OCR execution language group is acquired, and a character string search is executed.

  The search result transmission unit 306 transmits the search result to the user terminal 500 that has transmitted the search instruction (S306). The search result may be document image data including a search keyword, or may be text data after character conversion including the search keyword. Further, the display form of the portion corresponding to the search keyword in the document image data or text data may be changed. For example, the corresponding portion may be changed to a display form such as reverse display, color display, bold display, or framed display.

  As described above, the search server 300 executes the search process. According to the above configuration, the search accuracy can be improved because the language search according to the search condition is performed on the result of the OCR process with high character recognition accuracy.

[Modification]
The present embodiment is not limited to the above configuration. For example, the information processing system 10 may include a function for generating a text string searchable document file (for example, searchable PDF) as an OCR result and a function for causing the user terminal 500 to download the generated document file. In order to realize this function, the information processing system 10 generates a document file for each OCR execution language group, and manages the OCR target document in association with the generated document file and the OCR execution language group. And means for temporarily holding language information including keywords as search conditions when displaying the search result document, and for holding the download when the document file is instructed after the search result is displayed. And a means for downloading a document file associated with the language information. As a result, an appropriate character string searchable document desired by the user can be downloaded without any burden on the user.

  In the above description, a plurality of “languages” which are different from each other are given as examples. However, the information processing apparatus according to the present embodiment is not limited to “languages”. The information processing apparatus according to the present embodiment, for example, includes “Hiragana” and “Katakana” in the document to be read, “Hiragana” and “Kanji”, or “Hiragana” and “ The present invention is applicable even when the character types (character sets) are different from each other, such as when “Arabic numerals” are included. The character set includes, for example, character types such as hiragana, katakana, kanji, and Arabic numerals in addition to languages. Also, different character sets of the same character type or the same language may be used as different character sets. For example, in a document including “Hiragana” (first character set) and “Katakana” (second character set), the OCR execution unit 203 performs OCR processing on the hiragana portion using Hiragana (OCR execution language). The katakana portion is subjected to OCR processing using katakana (OCR execution language). As a result, the hiragana part is reliably converted to hiragana, and the katakana part is reliably converted to katakana.

  As described above, the information processing apparatus 10 according to the present embodiment includes the data receiving unit 201 that acquires document image data obtained by reading a document including a plurality of different character sets as an image, and the character for each of the plurality of character sets. An OCR execution unit 203 for recognizing the entire document image data by the set and converting the document image data into document data, and generating a number of the document data corresponding to the number of the character sets for one document image data; A search processing unit 305 that executes a search for input characters input as a search condition in at least one of the document data generated by the OCR execution unit 203.

  DESCRIPTION OF SYMBOLS 10 Information processing system, 100 Information processing apparatus, 200 OCR server, 300 Search server, 400 MFP, 500 User terminal, 21,31 CPU, 22,32 Memory, 23,33 Storage part, 24,34 Communication part, 201 Data Receiving unit, 202 OCR execution language determination unit, 203 OCR execution unit, 204 OCR result transmission unit, 205 data storage unit, 206 language information storage unit, 207 OCR result storage unit, 301 OCR result reception unit, 302 OCR information management unit, 303 Search instruction receiving unit 304 Search language determination unit 305 Search processing unit 306 Search result transmission unit 307 OCR result storage unit 308 OCR information storage unit 309 Character information storage unit 309

Claims (7)

Data acquisition means for acquiring document image data obtained by reading a document including a plurality of different character sets as an image;
For each of the plurality of character sets, the entire document image data is recognized by the character set and converted into document data, and the number of the documents corresponding to the number of the character sets for one document image data. Document data generating means for generating data;
Document data storage means for storing a plurality of the document data generated by the document data generation means ;
A search character set determination means for determining a character set of an input character input as a search condition;
Among the plurality of document data stored in the document data storage means, for the document data corresponding to the determined character set of the input character by the search character set determining unit, entered as the search condition a search executing means for executing a search of the input character,
An information processing apparatus comprising: If the document includes different first and second character sets,
The document data generation means recognizes the entire document image data based on the first character set and converts it into document data to generate first document data, and based on the second character set. Recognizing the entire document image data and converting it to document data to generate second document data;
The information processing apparatus according to claim 1. If the document includes different first and second character sets,
The document data generation means recognizes the entire document image data based on the first character set and the third character set whose character shapes are dissimilar to each other, converts the document image data into document data, and converts the first document set into document data. The second document data is generated by generating the data, recognizing the entire document image data based on the second character set and the fourth character set having dissimilar character shapes, and converting the document image data into document data. Generate
The information processing apparatus according to claim 1. The document data corresponding to the character set of the inputted input character as the search condition further comprises document data obtaining means for obtaining from the document data storage unit,
When the character set of the inputted input character as the search condition is a fifth character set,
The document data acquisition means acquires at least one document data including the fifth character set from the document data storage means;
The search executing means, for each said were acquisition document data, perform a search of the input character,
The information processing apparatus according to claim 1. The plurality of different character sets are a plurality of different languages.
The information processing apparatus according to claim 1. An information processing system including a character conversion device and a search device,
The character conversion device includes:
Data acquisition means for acquiring document image data obtained by reading a document including a plurality of different character sets as an image;
For each of the plurality of character sets, the entire document image data is recognized by the character set and converted into document data, and the number of the documents corresponding to the number of the character sets for one document image data. Document data generating means for generating data;
Document data storage means for storing a plurality of the document data generated by the document data generation means ;
Transmitting means for transmitting the document data to the search device,
The search device includes:
Receiving means for receiving the document data from the character conversion device;
A search character set determination means for determining a character set of an input character input as a search condition;
Among the plurality of document data stored in the document data storage means, for the document data corresponding to the determined character set of the input character by the search character set determining unit, entered as the search condition a search executing means for executing a search of the input character,
Transmitting means for transmitting a search result by the search execution means to a user terminal of a user who has input the search condition;
An information processing system comprising: Data acquisition means for acquiring document image data obtained by reading a document including a plurality of different character sets as an image;
For each of the plurality of character sets, the entire document image data is recognized by the character set and converted into document data, and the number of the documents corresponding to the number of the character sets for one document image data. Document data generating means for generating data ,
Document data storage means for storing a plurality of the document data generated by the previous SL document data generating means,
Search character set determination means for determining a character set of input characters input as a search condition, and
Among the plurality of document data stored in the document data storage means, for the document data corresponding to the determined character set of the input character by the search character set determining unit, entered as the search condition search execution means for executing a search of the input character,
As a program to make the computer function as.

Images