JP2020123273A - Information processing device and information processing program - Google Patents

Abstract

To allow for accurately identifying character strings contained in an image without referring to a predefined reference list.SOLUTION: An information processing device 10 is provided, comprising: an acquisition unit 22 configured to acquire character strings recognized from an image by OCR processing; an extraction unit 24 configured to extract character string candidates from the character strings acquired by the acquisition unit 22 using an extraction pattern expressed in regular expression; a correction unit 26 configured to correct the character string candidates using false recognition patterns in which the OCR processing tends to make false recognition; and an output unit 28 configured to output character string candidates identified from among the character string candidates corrected by the correction unit 26 using the regular expression.SELECTED DRAWING: Figure 2

Description

The present invention relates to an information processing device and an information processing program.

In Patent Document 1, in an information retrieval device that retrieves predetermined information from information retention means that retains a plurality of information, a normalization level is determined from a document image input means, a character recognition means, and a certainty factor of character recognition. Means, a normalization means for replacing with a representative character based on the normalization level, a search means for searching the information holding means for information including similar texts, and a search result output means for outputting the search result. An information search device characterized by the above is disclosed.

Patent Document 2 is an information processing apparatus that searches a character string for a specific term included in a given dictionary, and uses an n-gram method to perform partial characters of the character string for the given dictionary. Extraction means for performing a column search and extracting one or more specific term candidates included in the character string from the given dictionary, and a DP matching method for each of the one or more specific term candidates are used. Deriving means for deriving an editing distance between a specific term candidate and a partial character string in the character string corresponding to the specific term candidate, and the one or more specific term candidates in which the editing distance is shorter than a predetermined threshold value. And an output unit that outputs a partial character string in the character string.

JP, 2007-323415, A JP, 2008-81451, A

It is an object of the present invention to provide an information processing device and an information processing program capable of accurately specifying a character string included in an image without using a predetermined reference list.

An information processing apparatus according to a first aspect uses an extraction unit that acquires a character string recognized from an image by a character recognition process and an extraction pattern represented by a regular expression from the character string acquired by the acquisition unit. An extraction unit that extracts a character string candidate, a correction unit that corrects the character string candidate using an erroneous recognition pattern that is erroneously recognized by the character recognition process, and a character that is corrected by the correction unit using the regular expression. An output unit that outputs the character string candidates specified from the column candidates.

An information processing apparatus according to a second aspect is the information processing apparatus according to the first aspect, wherein the extraction unit uses a modified extraction pattern represented by a modified regular expression obtained by modifying the regular expression using the misrecognition pattern. To extract the character string candidates.

An information processing apparatus according to a third aspect is the information processing apparatus according to the first or second aspect, wherein the acquisition unit acquires the image, and the correction unit performs character recognition of the image targeting a specific character. The output unit includes a character recognition processing unit that performs processing, and a generation unit that generates a misrecognized character string candidate that is a character string candidate of the same type as the character string recognized by the character recognition processing unit. The specified character string candidate is output from the recognized character string candidates.

An information processing apparatus according to a fourth aspect is the information processing apparatus according to the third aspect, wherein the generation unit adds a character string generated using an erroneous recognition pattern designated by a user to the erroneously recognized character string candidates. ..

An information processing apparatus according to a fifth aspect is the information processing apparatus according to the third aspect or the fourth aspect, wherein the generation unit generates the misrecognized character string candidates except for the character string candidates represented by the extraction pattern. To do.

An information processing apparatus according to a sixth aspect is the information processing apparatus according to any one of the third to fifth aspects, wherein the character recognition processing unit calculates a score indicating accuracy of character recognition, and the output unit is If the score is equal to or greater than a predetermined threshold value, the specified character string candidate is output.

An information processing apparatus according to a seventh aspect is the information processing apparatus according to any one of the third to fifth aspects, wherein the character recognition processing unit calculates a score indicating accuracy of character recognition, and the output unit is If the score is less than a predetermined threshold value, the specified character string candidate is output as an estimated character string candidate.

An information processing apparatus according to an eighth aspect is the information processing apparatus according to any one of the third to fifth aspects, wherein the character recognition processing unit calculates a score indicating accuracy of character recognition, and the output unit is When the score is equal to or higher than a predetermined threshold and the specified character string candidates are plural, the plural specified character string candidates are output as estimated character string candidates.

An information processing apparatus according to a ninth aspect is the information processing apparatus according to any one of the third to fifth aspects, in which a necessary character string that is a necessary character string from a character string recognized by the character recognition processing unit. The generating unit further includes a necessary character string extracting unit that extracts a character string candidate that is a character string candidate of the same type as the necessary character string extracted by the necessary character string extracting unit.

An information processing apparatus according to a tenth aspect is the information processing apparatus according to any one of the first to fifth aspects, wherein the output unit is corrected by the correction unit to a character string represented by the regular expression. When there is no character string candidate, a character string that is close to the character string candidate corrected by the correction unit among the character strings represented by the regular expression is output as an estimated character string candidate.

An information processing apparatus according to an eleventh aspect is the information processing apparatus according to the tenth aspect, wherein the output unit is a character string that is different from the character string candidate corrected by the correction unit in the character string represented by the regular expression. The character string with the smallest number of is output as an estimated character string candidate.

The information processing program according to the twelfth aspect is an information processing program for causing a computer to function as each unit of the information processing apparatus according to any one of the first to eleventh aspects.

According to the first and twelfth aspects, there is an effect that a character string included in an image can be accurately specified without using a predetermined reference list.

According to the second aspect, as compared with the case of extracting the character string candidates without using the modified extraction pattern represented by the modified regular expression in which the regular expression is modified by using the misrecognition pattern, the character string that should be originally extracted It is possible to extract a character string in which some of the characters are erroneously recognized.

According to the third aspect, there is an effect that the character string included in the image can be specified more accurately than in the case where the erroneously recognized character string candidate is not generated.

According to the fourth aspect, it is possible to accurately specify the character string included in the image, as compared with the case where the misrecognized character string candidate is generated without using the misrecognized pattern specified by the user. Have.

According to the fifth aspect, it is possible to shorten the time until the character string included in the image is specified, as compared with the case where the misrecognized character string candidate is generated without removing the character string candidate represented by the extraction pattern. It has the effect that

According to the sixth aspect, the reliability that the specified character string candidate is output is increased as compared with the case where the specified character string candidate is output regardless of the score indicating the accuracy of character recognition. Have.

According to the seventh aspect, the reliability of being output as the estimated character string candidate is higher than that in the case of outputting the identified character string candidate as the estimated character string candidate regardless of the score indicating the accuracy of character recognition. It has the effect of increasing.

According to the eighth aspect, as compared with the case where a plurality of specified character string candidates are output as estimated character string candidates regardless of the score indicating the accuracy of character recognition, the output as a plurality of estimated character string candidates is performed. Has the effect of increasing the reliability of.

According to the ninth aspect, it is possible to appropriately generate the erroneously recognized character string candidate, as compared with the case where the erroneously recognized character string candidate is generated without extracting the necessary character string from the character string subjected to character recognition. Has the effect.

According to the tenth aspect, when there is no character string candidate in the character string represented by the regular expression, as compared with the case where nothing is output, the character string that may be included in the image can be grasped. Has the effect of.

According to the eleventh aspect, the reliability of the estimated character string candidate is increased as compared with the case where the estimated character string candidate is specified regardless of the number of different characters in the character string represented by the regular expression and the character string candidate. Has the effect of.

It is a block diagram of an image forming apparatus. FIG. 3 is a functional block diagram of a CPU according to the first embodiment. 3 is a flowchart of information processing according to the first embodiment. It is a figure which shows an example of the image which read the drawing. It is a figure for demonstrating the area|region used as the object of a re-recognition process. It is a figure which shows an example of misrecognized character information. It is a figure which shows an example of misrecognition pattern information. It is a functional block diagram of CPU which concerns on 2nd Embodiment. It is a flow chart of information processing concerning a 2nd embodiment.

Embodiments for carrying out the present invention will be described in detail below with reference to the drawings.

(First embodiment)

FIG. 1 is a configuration diagram of an image forming apparatus 1 according to the present embodiment. As shown in FIG. 1, the image forming apparatus 1 includes a scanner unit 2, an image forming unit 3, an operation unit 4, a display unit 5, a communication unit 6, a storage unit 7, and an information processing device 10. Note that, in FIG. 1, a supply mechanism that supplies a recording medium such as paper, a transport mechanism that transports the recording medium, and a discharge mechanism that discharges the recording medium are not shown.

The image forming apparatus 1 functions as a multifunction peripheral having a plurality of functions such as a scan function of reading a document, a copy function of copying a document, a print function of printing print data, and a facsimile transmission function of facsimile transmission of a document.

The scanner unit 2 optically reads a recording medium such as paper on which an image is formed, generates a read image, and outputs the read image to the information processing apparatus 10.

The image forming unit 3 has a function of forming an image on a recording medium such as paper by, for example, an electrophotographic system or an inkjet recording system, and forms an image on the recording medium according to an instruction from the information processing device 10.

The operation unit 4 is configured to include operation keys that receive various operations.

The display unit 5 is composed of a liquid crystal display or the like, and is composed of, for example, a touch panel that receives various operations by touching the screen with a finger or the like.

The communication unit 6 is an interface for performing data communication with an external device or the like.

The storage unit 7 is composed of a non-volatile storage device such as a hard disk, and stores an information processing program described later.

The information processing device 10 includes a CPU (Central Processing Unit) 10A, a ROM (Read Only Memory) 10B, a RAM (Random Access Memory) 10C, a nonvolatile memory 10D, and an input/output interface (I/O) 10E. The CPU 10A, the ROM 10B, the RAM 10C, the non-volatile memory 10D, and the I/O 10E are connected to each other via the bus 10F. The scanner unit 2, the image forming unit 3, the operation unit 4, the display unit 5, the communication unit 6, and the storage unit 7 are connected to the I/O 10E. The CPU 10A reads and executes the information processing program stored in the storage unit 7.

The image forming apparatus 1 has a drawing information extraction function in addition to the basic functions of the multi-function peripheral such as the copy function described above. The drawing information extraction function is a function of reading a drawing used in the manufacturing industry, equipment/construction industry, and the like, and extracting a specified character string described in the drawing as drawing information from an image of the read drawing. In the specified character string shown in the drawing, for example, various information such as basic attributes including product number, product name, and version, specification information showing the specifications of the product shown in the drawing, material information, design standard document number, etc. Contains a string that represents. The drawing information extracted from the drawing is printed on a recording medium such as paper, stored in the storage unit 7 as a data file, transmitted to an external device via the communication unit 6, or displayed on the display unit 5. ..

Cases in which the drawing information extraction function is used include the following cases. For example, if a character string shown in the drawing is written based on a certain technical standard and the technical standard needs to be updated and the character string shown in the drawing needs to be corrected, the old technical standard should be used. This is the case, for example, when searching for a drawing that includes the specified character string.

The functional configuration of the CPU 10A when the image forming apparatus 1 executes the drawing information extraction function will be described below.

As shown in FIG. 2, the CPU 10A functionally includes an OCR processing unit 20, an acquisition unit 22, an extraction unit 24, a correction unit 26, and an output unit 28.

The OCR processing unit 20 performs character recognition processing, that is, OCR (Optical Character Recognition) processing on the image read by the scanner unit 2 from the document, and outputs the character string included in the read image to the acquisition unit 22. The OCR processing executed by the OCR processing unit 20 is, for example, rule-based OCR processing. That is, all character strings included in the read image are extracted based on a predetermined rule.

The acquisition unit 22 acquires the character string recognized from the image by the OCR processing of the OCR processing unit 20. The acquisition unit 22 outputs the acquired character string to the correction unit 26.

The extraction unit 24 extracts character string candidates from the character string acquired by the acquisition unit 22, using an extraction pattern represented by a regular expression. It should be noted that the character string recognized by the OCR processing by the OCR processing unit 20 often includes a character that is erroneously recognized. Therefore, the extraction unit 24 extracts the character string candidates using the extraction pattern represented by the modified regular expression that is modified by using the misrecognized pattern that is misrecognized by the OCR processing of the OCR processing unit 20.

The correction unit 26 corrects the character string candidates extracted by the extraction unit 24 using the erroneous recognition pattern erroneously recognized by the OCR process.

Specifically, the correction unit 26 includes a re-recognition processing unit 30 and a generation unit 32.

The re-recognition processing unit 30 performs the OCR processing of the image again for the specific character. The re-recognition processing unit 30 executes the OCR processing of the image using machine learning such as a neural network. The re-recognition processing unit 30 is an example of a character recognition processing unit.

The generation unit 32 generates a misrecognized character string candidate that is a character string candidate of the same type as the character string candidate recognized by the re-recognition processing unit 30. For example, the misrecognized character string candidate is generated using the modified extraction pattern represented by the modified regular expression in which the regular expression is modified using the misrecognized pattern.

Here, the erroneously recognized character string candidate is a character string candidate recognized by the re-recognition processing unit 30 and is a character string candidate having the same type as the recognized character and difficult to recognize, or the re-recognition processing unit 30. The recognized character string is a character string candidate in which the recognized character is highly likely to be incorrect. In the following, it is assumed that the character string recognized by the re-recognition processing unit 30 is also included in the erroneously recognized character string candidates.

The output unit 28 outputs the character string candidates corrected by the correction unit 26, that is, the character string candidates identified from the misrecognized character string candidates generated by the generation unit 32, using the regular expression.

Next, the operation of the information processing device 10 according to the present embodiment will be described with reference to FIG. As shown in FIG. 2, the information processing program 40 is stored in the storage unit 7. The information processing shown in FIG. 3 is executed by the CPU 10A reading and executing the information processing program 40. The information processing illustrated in FIG. 3 is executed, for example, when the user sets a drawing on the scanner unit 2 and operates the operation unit 4 to instruct execution of the drawing information extraction process.

In step S100, the CPU 10A instructs the scanner unit 2 to read the drawing, and acquires the image of the drawing read by the scanner unit 2.

In step S102, the CPU 10A executes rule-based OCR processing on the image acquired in step S100. As a result, all the character strings included in the read image are acquired. Further, the positions of all the character strings included in the read image, that is, the coordinates in the image are also acquired. The character string recognized by the OCR process is obtained as a character string in the XML format, for example, but the format of the character string is not limited to the XML format.

In step S104, CPU 10A extracts a character string candidate included in the image acquired in step S100 from the character string obtained by the OCR processing in step S102. Specifically, the character string candidates included in the image are extracted using the extraction pattern represented by the regular expression. In the present embodiment, for example, as shown in FIG. 2, regular expression information 42 in which a regular expression is defined is stored in advance in the storage unit 7, and extraction represented by the regular expression defined in the regular expression information 42 is performed. Extract character string candidates using the pattern. The regular expression information 42 is not limited to being stored in the storage unit 7 in advance, and the regular expression information 42 may be acquired from an external device via the communication unit 6.

Here, the regular expression is a method of expressing a set of character strings by one character string. The regular expression corresponds to the character string that can be included in the drawing from which the drawing information is extracted. That is, the regular expression information 42 stored in advance in the storage unit 7 defines the regular expressions corresponding to the character strings that can be included in the drawing from which the drawing information is extracted.

For example, a character string that can be included in a drawing from which drawing information is extracted is composed of four characters, the first character is "A", the second character is "B", and the third character is "C". It is assumed that the fourth character is a character string in which any of the 10 types of numbers "0" to "9" is a character. In this case, the regular expression of the character string that can be included in the drawing from which the drawing information is extracted is represented by one character string “[A][B][C][0-9]”. In this case, the regular expression information 42 defines “[A][B][C][0-9]” as a regular expression. The character string of the extracted pattern represented by the regular expression “[A][B][C][0-9]” is 10 patterns from “ABC0” to “ABC9”.

By the way, in many cases, the character string recognized by the OCR processing in step S102 includes a character that is erroneously recognized. Therefore, in the above example, if the character string candidates are extracted using the extraction pattern represented by the regular expression “[A][B][C][0-9]”, A character string in which some of the characters are erroneously recognized will not be extracted. Therefore, in step S104, the character string candidates are extracted using the extraction pattern represented by the corrected regular expression corrected using the erroneous recognition pattern that is erroneously recognized in the OCR process. That is, the regular expression information 42 defines a modified regular expression that is corrected by using an erroneous recognition pattern that is erroneously recognized by the OCR process.

For example, in the case of the above example, “A” is likely to be erroneously recognized as “H”. Also, "B" is easily misrecognized as "8" or "day", for example. Further, "C" is apt to be erroneously recognized as "[" or "(", and "0" is apt to be erroneously recognized as "o" or "O". The modified regular expression modified using the recognition pattern is “[AH][B8 days][C[(][0oO-9]”. By using the extraction pattern represented by such modified regular expression By extracting the character string candidates, a character string in which some characters of the character string to be originally extracted are erroneously recognized is also extracted.

For example, assume that the image obtained in step S100 is the image 50 shown in FIG. The image 50 includes an image representing the character string “ABC0”, an image representing the character string “ABC8”, and an image representing the character string “XYZ1”.

In this case, since all the character strings included in the image 50 are recognized in the OCR processing of step S102, all of the character strings “ABC0”, the character string “ABC8”, and the character string “XYZ1” are all regarded as some character strings. Be recognized. Then, in the OCR processing of step S102, at least some characters may be erroneously recognized, so that not all character strings are recognized correctly. For example, the character string "ABC0" may be erroneously recognized as "ABCo" or "ABCO". Further, the character string "ABC8" may be erroneously recognized as "ABCB". On the other hand, in step S104, since the character string candidates are extracted using the extraction pattern represented by the modified regular expression, even if the character strings are erroneously recognized as "ABCo" or "ABCO", they are extracted as character string candidates. ..

In step S106, the CPU 10A executes an OCR process (hereinafter, referred to as a re-recognition process) using the learning model for the character string candidates extracted in step S104. That is, the re-recognition process is not executed for the entire image acquired in step S100, but the re-recognition process is executed only for the character string candidates extracted in step S104. Note that the target region for executing the re-recognition process may be set based on the position of the character string candidate obtained by the rule-based OCR process in step S102.

For example, as shown in FIG. 5, the re-recognition process is not performed on the entire image 50, but the re-recognition process is performed on the region R1 including the character string candidate “ABC0” and the region R2 including the character string candidate “ABC8”. To execute.

As an example of the OCR process using the learning model, an OCR process using machine learning such as a convolutional neural network (CNN) is used in the present embodiment, but the present invention is not limited to this.

The learning model has been learned for a particular character. Here, the specific character is a character that can be included in the drawing for which the drawing information extraction function is executed, and is, for example, an alphabet, a number, or a symbol.

As described above, the OCR processing using the learning model learned for a specific character has a higher character string recognition rate than the rule-based OCR processing executed in step S102.

For example, it is assumed that the character string “ABC0” included in the image 50 is erroneously recognized as “ABCo” and the character string “ABC8” is erroneously recognized as “ABCB” in the OCR process of step S102. Even in such a case, as a result of performing the re-recognition process on the regions R1 and R2 shown in FIG. 5 in the re-recognition process of step S106, “ABC0” and “ABC8” may be correctly recognized, respectively.

In the re-recognition process executed in step S106, a score representing the accuracy of character recognition is calculated. The higher the score, the higher the accuracy of character recognition, and the lower the score, the lower the accuracy of character recognition. For example, the score is a value in the range of 0 to 1, and the higher the numerical value, the higher the accuracy of character recognition.

The processing from step S108 below is executed for each character string recognized in the re-recognition processing.

In step S108, the CPU 10A generates a misrecognized character string candidate that is a character string candidate of the same type as the character string recognized by the re-recognition process of step S106.

Specifically, for example, as shown in FIG. 6, misrecognized character information 44 in which misrecognized characters are defined is stored in advance in the storage unit 7, and the misrecognized character string candidate is used by using this misrecognized character information 44. To generate. The misrecognized character information 44 defines a misrecognized pattern that is misrecognized in the OCR process of step S102. In the example of FIG. 6, “H” that is easily misrecognized for “A” is defined. Further, “8” is defined which is easily misrecognized for “B”. Further, “o” and “O” that are easily misrecognized with respect to “0” are defined.

Thereby, for example, when the character string recognized by the re-recognition process of step S106 is "ABC0", "HBC0", "A8C0", "ABCo", "ABCO", "H8C0", "H8Co", " H8CO" or the like is generated as a misrecognized character string candidate.

In step S110, the CPU 10A adds the character string generated using the erroneous recognition pattern designated by the user to the erroneously recognized character string candidates. Here, the erroneous recognition pattern designated by the user is, for example, an erroneous recognition pattern set according to the font of characters included in the drawing. For example, typical examples of characters that are easily erroneously recognized as "0" are "o" and "O", but depending on the font of the characters used in the drawings, characters other than "o" and "O" may be used. It may be easily misrecognized as another character. In this case, for example, as shown in FIG. 7, misrecognition pattern information 46 defining a user-specified misrecognition pattern is stored in the storage unit 7 in advance, and a character string generated using this misrecognition pattern information 46 is stored. Add to the misrecognized character string candidates.

In the example of the misrecognized pattern information 46 of FIG. 7, "day" is set as a character that is easily misrecognized for "B", and "D" is set as a character that is easily misrecognized for "0". For example, when the character string recognized by the re-recognition process in step S106 is "ABC0", "A date C0" and "ABCD" are added as the misrecognized character string candidates.

Note that among the misrecognized character string candidates generated in step S108 and step S110, the misrecognized character string candidates other than the character string represented by the regular expression are not necessary for specifying the character string candidates included in the image. is there.

Therefore, in step S112, the CPU 10A excludes the misrecognized character string candidates other than the character string of the extraction pattern represented by the regular expression from the misrecognized character string candidates generated in step S108 and step S110.

For example, in the above example, when the regular expression is “[A][B][C][0-9]”, “HBC0” and “A8C0” are set as the misrecognized character string candidates generated in step S108 and step S110. , "ABCo", "ABCO", "H8C0", "H8Co", "H8CO", "A day C0", "ABCD" are included. Since the character string represented by the regular expression is the alphabet from the first to the third character, and the fourth character from the first is a number, "ABCo" in which the fourth character from the first is alphabet among the above misrecognized character string candidates, “ABCO”, “H8Co”, “H8CO”, “ABCD”, and “A day C0” in which the second character from the beginning is a Chinese character are excluded from the misrecognized character string candidates.

In step S114, the CPU 10A identifies a character string candidate from the misrecognized character string candidates using the regular expression. That is, among the misrecognized character string candidates, the misrecognized character string candidates included in the character string represented by the regular expression are specified as the character string candidates. For example, when the regular expression is “[A][B][C][0-9]”, the character string represented by the regular expression is 10 patterns from “ABC0” to “ABC9”. Therefore, among the misrecognized character string candidates, the misrecognized character string candidates that match any of the character strings “ABC0” to “ABC9” are specified as the character string candidates. On the other hand, if the character string represented by the regular expression does not include any erroneously recognized character string candidates, the character string candidates are not specified.

In step S116, the CPU 10A determines whether the character string candidate is identified in step S114. Then, if the character string candidate is specified, that is, if there is a misrecognized character string candidate that matches any of the character strings represented by the regular expression, the process proceeds to step S118. On the other hand, if no character string candidate is specified, that is, if there is no misrecognized character string candidate that matches the character string represented by the regular expression, the process proceeds to step S124.

In step S118, the CPU 10A determines whether or not the score calculated in the re-recognition process of step S106 is equal to or higher than a predetermined threshold value. It should be noted that the threshold value is set to a value at which the reliability of the recognition result of the OCR process is equal to or higher than the allowable level if the score is equal to or higher than the threshold value. For example, the score is set to 90% or more of the maximum possible values. Specifically, when the range of values that the score can take is 0 to 1, the threshold is set to 0.95 as an example.

Then, when the score calculated in the re-recognition process of step S106 is equal to or higher than a predetermined threshold, that is, when the reliability of the recognition result of the re-recognition process of step S106 is equal to or higher than the allowable level, the process proceeds to step S119. On the other hand, if the score calculated in the re-recognition process of step S106 is less than the predetermined threshold, that is, if the reliability of the recognition result of the re-recognition process of step S106 is less than the allowable level, the process proceeds to step S122.

In step S119, it is determined whether the number of character string candidates identified in step S114 is one. Then, if there is one identified character string candidate, the process proceeds to step S120, and if there are multiple identified character string candidates, the process proceeds to step S122.

In step S120, CPU 10A outputs the one character string candidate identified in step S114. That is, the specified character string candidates are printed on a recording medium such as paper, stored in the storage unit 7 as a data file, transmitted to an external device via the communication unit 6, or displayed on the display unit 5. ..

On the other hand, step S122 is executed when the reliability of the recognition result of the re-recognition processing is determined to be less than the allowable level in step S118 or when a plurality of character string candidates are specified in step S119. Therefore, the character string candidates identified in step S114 are output as estimated character string candidates that are character string candidates estimated to be included in the drawing.

Further, step S124 is executed when there is no erroneously recognized character string candidate that matches the character string represented by the regular expression. Therefore, a character string close to the erroneously recognized character string candidate is searched from the character string represented by the regular expression. Specifically, the character string represented by the regular expression is searched for the character string having the smallest number of characters different from the erroneously recognized character string candidate. Then, the retrieved character string is output as an estimated character string candidate.

In step S126, it is determined whether or not the processes of steps S108 to S124 have been executed for all the character strings recognized in step S106. Then, when the processes of steps S108 to S124 have been executed for all the character strings recognized in step S106, this routine ends. On the other hand, if the processes of steps S108 to S124 have not been executed for all the character strings recognized in step S106, the process proceeds to step S108, and the processes of steps S108 to S124 are performed until the processing of all the character strings is completed. repeat.

As described above, in this embodiment, character string candidates are extracted from the character string recognized from the image by the OCR process using the extraction pattern represented by the regular expression. Then, the OCR process corrects the character string candidates using the erroneous recognition pattern, and the regular expression is used to output the character string candidates identified from the corrected character string candidates.

In the first embodiment, in step S110 of FIG. 3, the character string generated using the misrecognized pattern information 46 defining the user-specified misrecognized pattern is added to the misrecognized character string candidates. The process of S110 may be omitted.

Next, a specific example of specifying a character string candidate will be described.

・Specific example 1

In Example 1, the regular expression is [F][X][1][0][AZ][AZ][0-9][0-9][0-9], the actual string included in the drawing. Is "FX10AA106".

When the OCR process was executed under the above conditions, the recognized character string candidate was "FXIOAAIO6". That is, the third and fourth characters "10" are mistakenly recognized as "IO", and the seventh and eighth characters "10" are mistakenly recognized as "IO".

Next, the result of the re-recognition process was “FX10AA106”, which completely matched the actual character string included in the drawing. The score of the re-recognition process was equal to or higher than the threshold. In addition, as the erroneously recognized character string candidates, “FX10AA106”, “FX1OAA106”, “FX1oAA06”, “FX10AA1O6”, “FX10AA1o6”, etc. were generated including the character strings recognized by the re-recognition process. In addition, in the specific example 1, the process of further generating the misrecognized character string candidate by using the misrecognized pattern information designated by the user is not executed.

If unnecessary character string candidates are excluded from the misrecognized character string candidates by using a regular expression, only "FX10AA106" remains. This erroneously recognized character string candidate is a character string represented by a regular expression. Therefore, “FX10AA106” is specified and output as the character string candidate described in the drawing. This string exactly matches the actual string contained in the drawing.

・Specific example 2

The specific example 2 is the same as the specific example 1 except that the score of the re-recognition process is less than the threshold value.

In the second specific example, as in the first specific example, “FX10AA106” that exactly matches the actual character string included in the drawing is specified as the character string candidate described in the drawing, but the score of the re-recognition process is Since it is less than the threshold value, the reliability is low. Therefore, the character string candidate “FX10AA106” is output as an estimated character string candidate which is a character string candidate estimated to be included in the drawing.

・Specific example 3

In Example 3, the regular expression is [F][X][1][0][AZ][AZ][0-9][0-9][0-9,AZ] The character string is "FX10AA106".

When the OCR process was executed under the above conditions, the recognized character string candidate was "FXIOAAIOG". That is, the third and fourth characters “10” are erroneously recognized as “IO”, and the seventh to ninth characters “106” are erroneously recognized as “IOG”.

Next, the result of the re-recognition process is "FX10AA10G", and the ninth character "6" is erroneously recognized as "G". The score of the re-recognition process was equal to or higher than the threshold.
In addition, as the erroneously recognized character string candidates, "FX10AA10G", "FX1OAA10G", "FX1oAA0G", "FX10AA1OG", "FX10AA1oG", etc. were generated, including the character strings recognized by the re-recognition process.

Moreover, in the specific example 3, the character string generated using the user-specified misrecognition pattern information is added to the misrecognized character string candidates. In the user-specified misrecognition pattern information, "6:G" is defined as the misrecognition pattern. Therefore, "FX10AA106" and "FX10AA10G" are generated and added as the misrecognized character string candidates.

If unnecessary character string candidates are excluded from the misrecognized character string candidates using regular expressions, "FX10AA106" and "FX10AA10G" remain. Since all of these erroneously recognized character string candidates are character strings represented by regular expressions, they are output as estimated character string candidates.

(Second embodiment)

Next, a second embodiment will be described. The same parts as those in the first embodiment are designated by the same reference numerals, and detailed description thereof will be omitted. In the second embodiment, a case will be described in which the information to be extracted from the drawing is a character string representing the material of the plastic part.

The method of displaying the material of plastic parts is specified by JIS (Japanese Industrial Standard). For example, in the display of a single polymer or the like, an abbreviation of a material defined by JIS K 6899-1 is to be sandwiched by boundary characters ">" and "<". For example, when the material is acrylonitrile-butadiene-styrene and this abbreviation is ABS, it is written as ">ABS<".

In addition, in the display of the material of the plastic parts, in addition to the material abbreviation, "-", "+", "(", ")", etc. are used as delimiters, and the mass fraction or the content rate of the material, etc. A number that represents is used. For example, the expression ">PA66-(GF25+MD15)<" indicates polyamide 66 containing a mixture of 25% by mass of glass fiber (GF) and 15% by mass of mineral powder (MD).

All that is needed is a string containing the material abbreviations, and no delimiters. Therefore, in the second embodiment, for the character string sandwiched by the boundary characters, the character string excluding the boundary characters is extracted as the necessary character string.

FIG. 8 shows a functional block diagram of the CPU 10A according to the second embodiment. The functional block diagram shown in FIG. 8 is different from the functional block diagram shown in FIG. 2 in that a necessary character string extraction unit 31 is provided.

The required character string extraction unit 31 extracts a required character string from the character string recognized by the re-recognition processing by the re-recognition processing unit 30. Specifically, it is determined whether the character string recognized by the re-recognition process includes a boundary character. Then, when the character string recognized by the re-recognition processing includes a boundary character, the character string excluding the boundary character is extracted as a necessary character string.

Next, the operation of the information processing device 10 according to the present embodiment will be described with reference to FIG. The information processing shown in FIG. 9 is different from the information processing shown in FIG. 3 in that the processing of step S107 is added.

In step S107, a process of extracting a necessary character string is performed for each of the character strings recognized by the re-recognition process of step S106. That is, of the character strings recognized by the re-recognition process, the character string sandwiched between the boundary characters “>” and “<” is extracted. Then, for the character string sandwiched between the boundary characters “>” and “<”, the character string excluding the boundary character is extracted as a required character string. For example, when the character string recognized in the re-recognition process is “>ABS<”, “ABS” excluding the boundary character is extracted as the necessary character string. If the character string recognized in the re-recognition process is ">PA66-(GF25+MD15)<", "PA66-(GF25+MD15)" excluding the boundary character is extracted as the necessary character string.

Next, a specific example of specifying a character string candidate will be described.

・Specific example 1

In Specific Example 1, the regular expression is [>][][][][][][][][][][][][][][][][][][][][][ ][][][][<], the actual string contained in the drawing is “>PA66-(GF25+MD15)<”.

When the OCR process was executed under the above conditions, the recognized character string candidate was ">PAGG-(GF2S+MDI5)<". That is, the fourth and fifth characters "66" are mistakenly recognized as "GG", the tenth character "5" as "S", and the fourteenth character "1" as "I".

Next, the result of the re-recognition process was “>PA66-(GF25+MD15)<”, and the score of the re-recognition process was not less than the threshold. This string matches the string represented by the regular expression. Therefore, "PA66-(GF25+MD15)" is specified and output as the character string candidate described in the drawing.

・Specific example 2

The specific example 2 is the same as the specific example 1 except that the score of the re-recognition process is less than the threshold value. In this case, since the reliability of the recognition result of the re-recognition process is low, "PA66-(GF25+MD15)" is output as the estimated character string candidate.

Although the present invention has been described above using each embodiment, the present invention is not limited to the scope described in each embodiment. Various modifications and improvements can be added to each embodiment without departing from the gist of the present invention, and the modes to which the modifications or improvements are added are also included in the technical scope of the present invention.

For example, in each of the above-described embodiments, the case where the information processing apparatus is applied to the multifunction peripheral has been described as an example, but the information processing apparatus may be applied to a smartphone, a tablet terminal, or the like. In this case, a drawing may be taken with a camera mounted on a smartphone or a tablet terminal, and the process of FIG. 3 or 9 may be executed on the taken image.

Further, for example, the information processing shown in FIGS. 3 and 9 may be realized by hardware such as an ASIC (Application Specific Integrated Circuit). In this case, the processing speed can be increased as compared with the case of being realized by software.

Further, in each of the embodiments, the mode in which the information processing program is installed in the storage unit 7 has been described, but the invention is not limited to this. The information processing program according to the present embodiment may be provided in a form recorded in a computer-readable storage medium. For example, the information processing program according to the present embodiment is recorded in an optical disc such as a CD (Compact Disc)-ROM and a DVD (Digital Versatile Disc)-ROM, or a semiconductor such as a USB (Universal Serial Bus) memory and a memory card. You may provide in the form recorded on memory. Further, the information processing program according to the present embodiment may be acquired from an external device via a communication line connected to the communication unit 6.

1 image forming apparatus 2 scanner section 3 image forming section 4 operation section 5 display section 6 communication section 7 storage section 10 information processing apparatus 20 OCR processing section 22 acquisition section 24 extraction section 26 correction section 28 output section 30 re-recognition processing section 31 required Character string extraction unit 32 Generation unit 40 Information processing program 42 Regular expression information 44 False recognition character information 46 False recognition pattern information 50 Image

Claims (12)

An acquisition unit that acquires a character string recognized from an image by character recognition processing,
From the character string acquired by the acquisition unit, an extraction unit that extracts character string candidates using an extraction pattern represented by a regular expression,
A correction unit that corrects the character string candidates using an erroneous recognition pattern that the character recognition process erroneously recognizes,
An output unit that outputs the character string candidates identified from the character string candidates corrected by the correction unit using the regular expression,
Information processing device equipped with. The information processing apparatus according to claim 1, wherein the extraction unit extracts the character string candidates using a modified extraction pattern represented by a modified regular expression obtained by modifying the regular expression using the misrecognition pattern. The acquisition unit acquires the image,
The correction unit, a character recognition processing unit for performing character recognition processing of the image for a specific character,
A generation unit that generates a misrecognized character string candidate that is a character string candidate of the same type as the character string candidate recognized by the character recognition processing unit;
Including
The information processing apparatus according to claim 1, wherein the output unit outputs a character string candidate specified from the misrecognized character string candidates. The information processing apparatus according to claim 3, wherein the generation unit adds a character string generated using an erroneous recognition pattern designated by a user to the erroneously recognized character string candidate. The information processing apparatus according to claim 3, wherein the generation unit generates the misrecognized character string candidate by excluding the character string candidate represented by the extraction pattern. The character recognition processing unit calculates a score representing the accuracy of character recognition,
The information processing apparatus according to claim 3, wherein the output unit outputs the specified character string candidate when the score is equal to or higher than a predetermined threshold value. The character recognition processing unit calculates a score representing the accuracy of character recognition,
The information processing apparatus according to claim 3, wherein the output unit outputs the specified character string candidate as an estimated character string candidate when the score is less than a predetermined threshold value. The character recognition processing unit calculates a score representing the accuracy of character recognition,
The output unit outputs a plurality of the specified character string candidates as an estimated character string candidate when the score is equal to or more than a predetermined threshold and the specified character string candidates are plural. The information processing apparatus according to item 1. Further comprising a necessary character string extraction unit for extracting a necessary character string that is a necessary character string from the character string recognized by the character recognition processing unit,
The information processing according to any one of claims 3 to 5, wherein the generation unit generates a misrecognized character string candidate that is a character string candidate having the same type as the necessary character string extracted by the necessary character string extraction unit. apparatus. When the character string represented by the regular expression does not have a character string candidate modified by the modifying unit, the output unit includes a character string modified by the modifying unit among the character strings represented by the regular expression. The information processing apparatus according to claim 1, wherein a character string close to a column candidate is output as an estimated character string candidate. The output unit outputs, as an estimated character string candidate, a character string having the smallest number of characters different from the character string candidate corrected by the correction unit among the character strings represented by the regular expression. Information processing device. An information processing program for causing a computer to function as each unit of the information processing apparatus according to claim 1.