Information processing device, processing method, and processing program

Abstract

An information processing device (100) is provided with: an acquisition unit (120) that acquires an image (200) that includes a first symbol and a plurality of character strings, and a relationship table (111) that shows correspondence relationships between the symbols and the character strings; an extraction unit (130) that extracts a first symbol and a plurality of character strings on the basis of the image (200); a distance calculation unit (141) that calculates the distance between the first symbol and the plurality of character strings on the basis of the position of the first symbol and the respective positions of the plurality of character strings; and a determination unit (142) that determines a character string having a relationship with the first symbol on the basis of the relationship table (111) and the distance.

Description

Technical Field

[0001] The present disclosure relates to an information processing device, a processing method, and a processing program. Background Art

[0002] An image analysis device is known. The image analysis device analyzes an image. For example, the image to be analyzed contains symbols and character strings. A technology for extracting symbols and character strings is known (see Patent Document 1 or Non-Patent Document 1).

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent No. 6076128

[0006] Non-patent literature

[0007] Non-patent document 1: Yoneyama Shogo, Hirano Kei, Okada Yasuhiro, "Discussion on extraction method of シンボル within the image", Comprehensive Conference of the Society of Electronics, Information and Communications Technology, March 2006

[0008] Non-patent literature 2: T.Hirano et.al, "Structural character recognition using simulated annealing" ICDAR1997, August 1997 Summary of the invention

[0009] Problem that the invention aims to solve

[0010] However, there are sometimes a plurality of character strings around a symbol. In this case, there is a problem that it is unclear which character string is associated with the symbol.

[0011] The purpose of the present disclosure is to determine the appropriate association.

[0012] Means used to solve problems

[0013] An information processing device according to an embodiment of the present disclosure is provided. The information processing device comprises: an acquisition unit that acquires an image including a first symbol and a plurality of character strings, and relationship information showing a correspondence between the symbol and the character strings; an extraction unit that extracts the first symbol and the plurality of character strings based on the image; a distance calculation unit that calculates the distance between the first symbol and the plurality of character strings based on the position of the first symbol and the positions of the plurality of character strings; and a determination unit that determines a character string having a relationship with the first symbol based on the relationship information and the distance.

[0014] Effects of the Invention

[0015] According to the present disclosure, it is possible to determine an appropriate association. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a diagram showing hardware included in the information processing device according to the first embodiment.

[0017] Figure 2 This is a block diagram showing the functions of the information processing device according to the first embodiment.

[0018] Figure 3 This is a diagram showing an example of an image according to the first embodiment.

[0019] Figure 4 This is a diagram showing an example of the extraction result in the first embodiment.

[0020] Figure 5 This is a diagram showing an example of a relationship table according to the first embodiment.

[0021] Figure 6 This is a block diagram showing the functions of the relevance determination unit according to the first embodiment.

[0022] Figure 7 This is a flowchart showing an example of processing executed by the information processing device according to Embodiment 1.

[0023] Figure 8 This is a block diagram showing the functions of the information processing device according to the second embodiment.

[0024] Fig. 9 This is a block diagram showing the functions of a relevance determination unit according to the second embodiment.

[0025] Fig.10 This is a flowchart showing an example of processing executed by the information processing device according to the second embodiment.

[0026] Fig.11 This is a block diagram showing the functions of the information processing device according to the third embodiment.

[0027] Fig.12 This is a diagram showing an example of correction information according to the third embodiment.

[0028] Fig.13 This is a flowchart showing an example of processing executed by the information processing device according to the third embodiment.

[0029] Fig.14 This is a block diagram showing the functions of the information processing device according to the fourth embodiment.

[0030] Fig.15 This is a diagram showing an example of correction information according to the fourth embodiment.

[0031] Fig.16This is a block diagram showing the functions of a relevance determination unit according to the fourth embodiment.

[0032] Fig.17 This is a diagram showing an example of an image according to the fourth embodiment.

[0033] Fig.18 This is a diagram showing an example of a list of determination results according to the fourth embodiment. DETAILED DESCRIPTION

[0034] Hereinafter, embodiments will be described with reference to the drawings. The following embodiments are merely examples, and various modifications can be made within the scope of the present disclosure.

[0035] Implementation method 1.

[0036] Figure 1 1 is a diagram showing hardware included in the information processing device according to Embodiment 1. The information processing device 100 is a device that executes the processing method. The information processing device 100 includes a processor 101 , a volatile storage device 102 , and a nonvolatile storage device 103 .

[0037] The processor 101 controls the entire information processing device 100. For example, the processor 101 is a CPU (Central Processing Unit), an FPGA (Field Programmable Gate Array), etc. The processor 101 may also be a multiprocessor. In addition, the information processing device 100 may also have a processing circuit.

[0038] The volatile storage device 102 is a main storage device of the information processing device 100. For example, the volatile storage device 102 is a RAM (Random Access Memory). The nonvolatile storage device 103 is an auxiliary storage device of the information processing device 100. For example, the nonvolatile storage device 103 is a HDD (Hard Disk Drive) or an SSD (Solid State Drive).

[0039] Furthermore, the information processing device 100 may also include a scanner or a camera.

[0040] Next, the functions of the information processing device 100 will be described.

[0041] Figure 2 This is a block diagram showing the functions of the information processing device according to Embodiment 1. The information processing device 100 includes a storage unit 110 , an acquisition unit 120 , an extraction unit 130 , and a relevance determination unit 140 .

[0042] The storage unit 110 may be realized as a storage area reserved in the volatile storage device 102 or the nonvolatile storage device 103 .

[0043] Part or all of the acquisition unit 120, the extraction unit 130, and the association determination unit 140 may also be implemented by a processing circuit. In addition, part or all of the acquisition unit 120, the extraction unit 130, and the association determination unit 140 may also be implemented as a module of a program executed by the processor 101. For example, the program executed by the processor 101 is also referred to as a processing program. For example, the processing program is recorded in a recording medium.

[0044] The acquisition unit 120 acquires an image. For example, when the information processing device 100 has a camera, the acquisition unit 120 acquires an image obtained by photographing an object with the camera. In addition, for example, the acquisition unit 120 acquires an image from an external device. In addition, the figure of the external device is omitted. Here, an image is shown.

[0045] Figure 3 2 is a diagram showing an example of an image according to Embodiment 1. An image 200 is an image acquired by the acquisition unit 120. The image 200 includes symbols 201 to 203 and character strings 204 to 206.

[0046] The extraction unit 130 extracts the symbols 201 to 203 and the character strings 204 to 206 based on the image 200. For example, the extraction unit 130 extracts the symbols 201 to 203 and the character strings 204 to 206 using the extraction method described in Patent Document 1 or Non-Patent Document 1. The extraction result is shown.

[0047] Figure 4 300 is a diagram showing an example of the extraction result of Embodiment 1. The extraction result 300 includes elements, starting point coordinates, width, height, and results. For example, the extraction result 300 shows that the starting point coordinates of symbol 1 (i.e., symbol 201) are (10, 10). For example, the extraction result 300 shows that the width of symbol 1 is "50" and the height of symbol 1 is "50". In addition, the extraction result 300 shows that the shape of symbol 1 is a quadrilateral.

[0048] In addition, the center coordinates of the symbol may be included in the extraction result 300 instead of the starting point coordinates. Here, the starting point coordinates, the center coordinates, etc. are referred to as position coordinates.

[0049] The acquisition unit 120 acquires the relationship table. For example, the acquisition unit 120 acquires the relationship table from the storage unit 110. Alternatively, the acquisition unit 120 acquires the relationship table from an external device. Here, the relationship table is shown.

[0050] Figure 51 is a diagram showing an example of a relationship table in Embodiment 1. For example, a relationship table 111 is stored in the storage unit 110. The relationship table 111 is also called relationship information. The relationship table 111 is information showing the correspondence between symbols and character strings. For example, the relationship table 111 shows that "quadrilateral" has a relationship with "box", "area", and "floor".

[0051] Next, the relevance determination unit 140 will be described.

[0052] Figure 6 This is a block diagram showing the functions of the relevance determination unit according to Embodiment 1. The relevance determination unit 140 includes a distance calculation unit 141 and a determination unit 142 .

[0053] The distance calculation unit 141 calculates the distance between one symbol and the plurality of character strings based on the position of the symbol and the positions of the plurality of character strings. Here, the symbol is also referred to as a first symbol.

[0054] For example, the distance L1 between the symbol 201 and the character string 204 is calculated by equation (1). Specifically, the distance calculation unit 141 calculates the distance L1 based on the position coordinates of the symbol 201 and the position coordinates of the character string 204 .

[0055] [Formula 1]

[0056]

[0057] The distance L2 between the symbol 201 and the character string 205 is calculated by equation (2).

[0058] [Formula 2]

[0059]

[0060] The distance L3 between the symbol 201 and the character string 206 is calculated by equation (3).

[0061] [Formula 3]

[0062]

[0063] The distance calculation unit 141 similarly calculates the distance between the symbol 202 and the character strings 204 to 206. The distance calculation unit 141 similarly calculates the distance between the symbol 203 and the character strings 204 to 206. The distance may be expressed by Euclidean distance or other representations.

[0064] The determination unit 142 determines the character string that has a relationship with the symbol based on the relationship table 111 and the distance. The following is an explanation of the case of determining the character string that has a relationship with the symbol 201. Here, the distance is used in order from small to large. According to the above calculation results, the distance L2 is the smallest. Therefore, it is considered that the character string 205 has a relationship with the symbol 201. The determination unit 142 refers to the relationship table 111. Based on the relationship table 111, the determination unit 142 determines that the symbol 201 (i.e., quadrilateral) and the character string 205 (i.e., tent) have no relationship. Then, the determination unit 142 selects the distance L1 as the second smallest distance. The determination unit 142 refers to the relationship table 111. Based on the relationship table 111, the determination unit 142 determines that the symbol 201 (i.e., quadrilateral) and the character string 204 (i.e., box) have a relationship.

[0065] The determination unit 142 similarly determines the character string related to the symbol 202. Furthermore, the determination unit 142 similarly determines the character string related to the symbol 203.

[0066] The relevance determination unit 140 outputs the determination result.

[0067] Next, the processing executed by the information processing device 100 will be described using a flowchart.

[0068] Figure 7 This is a flowchart showing an example of processing executed by the information processing device according to Embodiment 1.

[0069] (Step S11 ) The acquisition unit 120 acquires the image 200 .

[0070] (Step S12 ) The extraction unit 130 obtains the symbols 201 to 203 and the character strings 204 to 206 .

[0071] (Step S13) The distance calculation unit 141 calculates the distance between the symbol 201 and the character strings 204 to 206. The distance calculation unit 141 calculates the distance between the symbol 202 and the character strings 204 to 206. The distance calculation unit 141 calculates the distance between the symbol 203 and the character strings 204 to 206.

[0072] (Step S14) The determination unit 142 determines a character string related to the symbol 201 based on the relationship table 111 and the distance. The determination unit 142 determines a character string related to the symbol 202 based on the relationship table 111 and the distance. The determination unit 142 determines a character string related to the symbol 203 based on the relationship table 111 and the distance.

[0073] For example, character strings 204 to 206 exist around symbol 201. Information processing device 100 determines that symbol 201 has a relationship with character string 204. Therefore, according to Embodiment 1, information processing device 100 can determine an appropriate relationship.

[0074] In addition, the determination unit 142 determines the character string that has a relationship with the symbol based on the distance below the predetermined threshold value TH among the calculated multiple distances and the relationship table 111. In other words, the determination unit 142 does not use the distance greater than the threshold value TH among the multiple distances to determine whether there is a relationship with the symbol. In the case where the distance separation between the symbol and the character string is large, the symbol and the character string do not have a relationship. Therefore, the distance greater than the threshold value TH is excluded from the determination material. In addition, by excluding the distance greater than the threshold value TH, the number of times the determination unit 142 refers to the relationship table 111 is reduced. Therefore, the processing time is shortened.

[0075] Implementation method 2.

[0076] Next, Embodiment 2 will be described. In Embodiment 2, matters different from Embodiment 1 will be mainly described. In Embodiment 2, description of matters common to Embodiment 1 will be omitted.

[0077] Figure 8 Detailed description of the invention Figure 2 is a block diagram showing the functions of the information processing device according to the second embodiment. Figure 2 The same structure as shown Figure 8 The structure is annotated with Figure 2 The same reference numerals are shown.

[0078] The information processing device 100a includes an extraction unit 130a, an association determination unit 140a, and a character recognition unit 150. Part or all of the extraction unit 130a, the association determination unit 140a, and the character recognition unit 150 may also be implemented by a processing circuit. In addition, part or all of the extraction unit 130a, the association determination unit 140a, and the character recognition unit 150 may also be implemented as a module of a program executed by the processor 101.

[0079] The extracting unit 130a extracts the symbols 201 to 203 based on the image 200. Furthermore, the extracting unit 130a extracts the positions of the character strings 204 to 206. That is, the extracting unit 130a does not extract the character strings 204 to 206.

[0080] The character recognition unit 150 performs character recognition based on the image. Specifically, the character recognition unit 150 performs character recognition based on the positions of the character strings 204 to 206. For example, the character recognition unit 150 performs character recognition by the method described in non-patent document 2. Here, the character recognition unit 150 sometimes generates character recognition errors. For example, the character recognition unit 150 recognizes the character string "テント" (Chinese meaning: tent) as the character string "テソト".

[0081] Next, the relevance determination unit 140 a will be described.

[0082] Fig. 9 This is a block diagram showing the functions of the association determination unit according to Embodiment 2. The association determination unit 140a includes a determination unit 142a.

[0083] The determination unit 142a detects multiple strings that are the same as or similar to the multiple strings obtained through character recognition from the relationship table 111. The determination unit 142a determines the strings related to the symbol based on the detected multiple strings, the relationship table 111, and the distance.

[0084] Hereinafter, the case of determining the strings related to the symbol 203 will be specifically described. The strings "box", "テソト", and "life buoy" are obtained through character recognition. The determination unit 142a detects the string "box" that is the same as the string "box" from the relationship table 111. The determination unit 142a detects the string "tent" that is similar to the string "テソト" from the relationship table 111. For example, the determination unit 142a detects similar strings by performing string comparison. The determination unit 142a detects the string "life buoy" that is the same as the string "life buoy" from the relationship table 111. The determination unit 142a determines the strings related to the symbol 203 based on the detected strings "box", "tent", and "life buoy", the relationship table 111, and the distance. The distances are used in ascending order. For example, the determination unit 142a determines based on the relationship table 111 that the symbol 203 (i.e., the triangle) is related to "tent". The association determination unit 140 outputs a determination result indicating that the symbol 203 (i.e., the triangle) is related to "tent".

[0085] Next, a flowchart will be used to illustrate the processing performed by the information processing device 100a.

[0086] Fig.10 This is a flowchart showing an example of the processing performed by the information processing device according to Embodiment 2.

[0087] (Step S21) The acquisition unit 120 acquires the image 200.

[0088] (Step S22) The extraction unit 130a extracts the positions of the symbols 201 to 203 and the strings 204 to 206.

[0089] (Step S23) The character recognition unit 150 performs character recognition based on the image.

[0090] (Step S24) The distance calculation unit 141 calculates the distances between the symbol 201 and the strings 204 to 206. The distance calculation unit 141 calculates the distances between the symbol 202 and the strings 204 to 206. The distance calculation unit 141 calculates the distances between the symbol 203 and the strings 204 to 206.

[0091] (Step S25 ) The determination unit 142 a detects, from the relationship table 111 , a plurality of character strings that are identical to or similar to the plurality of character strings obtained by character recognition.

[0092] (Step S26) The determination unit 142a determines a character string related to the symbol 201 based on the detected plurality of character strings, the relationship table 111, and the distance. The determination unit 142a determines a character string related to the symbol 202 based on the detected plurality of character strings, the relationship table 111, and the distance. The determination unit 142a determines a character string related to the symbol 203 based on the detected plurality of character strings, the relationship table 111, and the distance.

[0093] According to the second embodiment, the information processing device 100a can detect a character string having a relationship with a symbol even when a character recognition error occurs.

[0094] Implementation method 3.

[0095] Next, Embodiment 3 will be described. In Embodiment 3, matters different from Embodiment 1 will be mainly described. In Embodiment 3, description of matters common to Embodiment 1 will be omitted.

[0096] Fig.11 Detailed description of the invention Figure 1 is a block diagram showing the functions of an information processing device according to Embodiment 3. Figure 2 The same structure as shown Fig.11 The structure is annotated with Figure 2 The information processing device 100 further includes a character recognition unit 160 and a correction unit 170 .

[0097] Part or all of the character recognition unit 160 and the correction unit 170 may also be implemented by a processing circuit. In addition, part or all of the character recognition unit 160 and the correction unit 170 may also be implemented as a module of a program executed by the processor 101 .

[0098] The character recognition unit 160 performs character recognition based on the image. For example, the character recognition unit 160 performs character recognition using the method described in Non-Patent Document 2.

[0099] The acquisition unit 120 acquires the correction information. For example, the acquisition unit 120 acquires the correction information from the storage unit 110. Also, for example, the acquisition unit 120 acquires the correction information from an external device. The correction information is specifically shown.

[0100] Fig.12 1 is a diagram showing an example of correction information according to Embodiment 3. For example, correction information 112 is stored in storage unit 110. Correction information 112 is information indicating a character string that is preferentially associated among a character string of a character recognition result and a character string of a determination result.

[0101] The correction unit 170 determines the character string to be associated based on the character string of the judgment result, the result of character recognition, i.e., the character recognition result, and the correction information 112. When the character string of the judgment result is different from the determined character string (hereinafter referred to as the determined character string), the correction unit 170 corrects the character string of the judgment result to the determined character string. For example, the character recognition result is "floor". The judgment result shows that "quadrilateral" and "box" have a relationship. That is, the character string of the judgment result is "box". The correction unit 170 determines the associated character string "floor" based on the character string "box" of the judgment result, the character recognition result "floor" and the correction information 112. The character string "box" of the judgment result is different from the determined character string "floor". Therefore, the correction unit 170 corrects the character string "box" of the judgment result to the determined character string "floor". As a result, information indicating that "quadrilateral" and "floor" have a relationship is output.

[0102] Next, the processing executed by the information processing device 100 will be described using a flowchart.

[0103] Fig.13 This is a flowchart showing an example of processing executed by the information processing device according to the third embodiment. Fig.13 The processing of steps S15 to S18 is the same as Figure 7 Therefore, in Fig.13 Steps S15 to 18 are described in the following. In addition, descriptions of processes other than steps S15 to 18 are omitted.

[0104] (Step S15 ) The character recognition unit 160 performs character recognition based on the image.

[0105] (Step S16 ) The correction unit 170 specifies the final character string based on the character recognition result and the correction information 112 .

[0106] (Step S17) The correction unit 170 determines whether the character string of the determination result is different from the determination character string. If they are different, the process proceeds to step S18. If they are the same, the information processing device 100 outputs the determination result.

[0107] (Step S18 ) The correction unit 170 corrects the character string of the determination result to a confirmed character string.

[0108] In addition, step S15 may be executed before step S12. In addition, step S15 may be executed in parallel with steps S12 to S14.

[0109] According to the third embodiment, information processing apparatus 100 can associate a character string to be associated with a symbol with priority.

[0110] In addition, the information processing device 100 may also perform the following processing. The acquisition unit 120 acquires the correct answer information of the character string. The correction unit 170 calculates the correct answer degree of the character string of the result of character recognition, i.e., the character recognition result, based on the correct answer information. In addition, the correction unit 170 calculates the correct answer degree of the character string of the determination result based on the correct answer information. The correction unit 170 compares the correct answer degree of the character string of the character recognition result with the correct answer degree of the character string of the determination result. When the correct answer degree of the character string of the character recognition result is greater than the correct answer degree of the character string of the determination result, the correction unit 170 corrects the character string of the determination result to the character string of the character recognition result.

[0111] The above-mentioned processing is explained in detail. The character recognition result is "floor". The judgment result indicates that "quadrilateral" and "box" are related. That is, the character string of the judgment result is "box". The correct answer information shows "floor". The correction unit 170 calculates the correct answer degree of the character string of the character recognition result based on the correct answer information. The correction unit 170 calculates the correct answer degree of the character string of the judgment result based on the correct answer information. The correction unit 170 determines that the correct answer degree of the character string of the character recognition result is greater than the correct answer degree of the character string of the judgment result. The correction unit 170 corrects the character string "box" of the judgment result to the character string "floor" of the character recognition result. Thus, information indicating that "quadrilateral" and "floor" are related is output.

[0112] In this way, even when the character string in the determination result is erroneous, the information processing device 100 can correct it to an appropriate character string.

[0113] Implementation method 4.

[0114] Next, Embodiment 4 will be described. In Embodiment 4, matters different from Embodiment 1 will be mainly described. In Embodiment 4, description of matters common to Embodiment 1 will be omitted.

[0115] Fig.14 4 is a block diagram showing the functions of the information processing device according to the fourth embodiment. Figure 2 The same structure as shown Fig.14 The structure is annotated with Figure 2 The information processing device 100 b includes a relevance determination unit 140 b and a determination correction unit 180 .

[0116] Part or all of the relevance determination unit 140 b and the determination correction unit 180 may be implemented by a processing circuit. In addition, part or all of the relevance determination unit 140 b and the determination correction unit 180 may be implemented as a module of a program executed by the processor 101 .

[0117] The acquisition unit 120 acquires the correction information. For example, the acquisition unit 120 acquires the correction information from the storage unit 110. Also, for example, the acquisition unit 120 acquires the correction information from an external device. The correction information is specifically shown.

[0118] Fig.15 1 is a diagram showing an example of correction information of Embodiment 4. For example, correction information 113 is stored in storage unit 110. Correction information 113 is information for correcting a determination result. Specifically, correction information 113 shows a relationship between an extracted character string and an extracted symbol. A specific method of using correction information 113 will be described later.

[0119] Next, the relevance determination unit 140b will be described.

[0120] Fig.16 This is a block diagram showing the functions of the relevance determination unit according to Embodiment 4. The relevance determination unit 140b includes a determination unit 142b.

[0121] The determination unit 142b determines the combination of the symbol and the character string closest to the symbol based on the calculated distance. The determination unit 142b determines whether the relationship between the symbol and the character string closest to the symbol is registered in the relationship table 111. If the relationship is registered in the relationship table 111, the determination unit 142b outputs the relationship as the determination result.

[0122] When the relationship is registered in the relationship table 111, the determination correction unit 180 determines whether to correct the determination result based on the correction information 113. When correcting the determination result, the determination correction unit 180 corrects the determination result based on the correction information 113.

[0123] Next, the processing executed by the information processing device 100b will be described using a specific example.

[0124] Fig.17 is a diagram showing an example of an image according to Embodiment 4. The determination unit 142b determines a combination of a symbol 201 and a character string 204 closest to the symbol 201 based on the calculated distance. Fig.17 In FIG. 14 , the combination is represented by combination 11. Combination 11 is also referred to as combination A. The determination unit 142 b determines that combination 11 is registered in the relationship table 111. Since the relationship indicated by the determination result is registered in the correction information 113, the determination correction unit 180 does not correct the determination result.

[0125] The determination unit 142b determines the combination of the symbol 203 and the character string 205 closest to the symbol 203 based on the calculated distance. Fig.17In FIG. 14 , the combination is represented by combination 12. Combination 12 is also referred to as combination B. The determination unit 142 b determines that combination 12 is registered in the relationship table 111. Since the relationship indicated by the determination result is registered in the correction information 113, the determination correction unit 180 does not correct the determination result.

[0126] The determination unit 142b determines the combination of the symbol 202 and the character string 206 that is closest to the symbol 202. Fig.17 , the combination is represented by combination 13. Combination 13 is also referred to as combination C. Here, the extraction unit 130 extracts symbol 202 as "quadrilateral". The determination unit 142b determines that combination 13 is registered in the relationship table 111. Since the relationship shown in the determination result is not registered in the correction information 113, the determination correction unit 180 corrects the determination result. Specifically, the determination correction unit 180 corrects the determination result "quadrilateral, life buoy" to "ellipse, life buoy" based on the correction information 113.

[0127] Fig.18 : is a diagram showing an example of a list of determination results in Embodiment 4. The determination result of combination A shows that "quadrilateral" and "box" are related. The determination result of combination B shows that "triangle" and "tent" are related. The determination result of combination C shows that "ellipse" and "life buoy" are related.

[0128] As described above, when the extraction unit 130 erroneously extracts the symbol 202 as a "quadrilateral", the determination correction unit 180 corrects the "quadrilateral" to an "ellipse". Thus, the information processing device 100b can correct the erroneously extracted symbol to a correct symbol. Furthermore, the information processing device 100b can output an accurate determination result.

[0129] The features of the embodiments described above can be combined with each other as appropriate.

[0130] Description of Reference Numerals

[0131] 100 information processing device, 100a information processing device, 100b information processing device, 101 processor, 102 volatile storage device, 103 non-volatile storage device, 110 storage unit, 111 relationship table, 112 correction information, 113 correction information, 120 acquisition unit, 130 extraction unit, 130a extraction unit, 140 association determination unit, 140a association determination unit, 140b association determination unit, 141 distance calculation unit, 142 determination unit, 142a determination unit, 142b determination unit, 150 character recognition unit, 160 character recognition unit, 170 correction unit, 180 determination and correction unit, 200 image, 201 to 203 symbols, 204 to 206 character strings, 300 extraction result.

Claims

1. An information processing device, wherein: The information processing device comprises: an acquisition unit that acquires an image including a first symbol and a plurality of character strings, and relationship information indicating a correspondence relationship between the symbol and the character strings; an extraction unit configured to extract the first symbol and the plurality of character strings based on the image; a distance calculation unit that calculates a distance between the first symbol and the plurality of character strings based on a position of the first symbol and positions of the plurality of character strings; as well as A determination unit determines a character string having a relationship with the first symbol based on the relationship information and the distance.

2. The information processing device according to claim 1, wherein: The determination unit determines a character string having a relationship with the first symbol based on a distance that is equal to or smaller than a predetermined threshold value among the calculated plurality of distances and the relationship information.

3. The information processing device according to claim 1 or 2, wherein: The information processing device further includes a character recognition unit that performs character recognition based on the image. The determination unit detects a plurality of character strings that are identical to or similar to the plurality of character strings obtained by the character recognition from the relationship information, and determines a character string that is related to the first symbol based on the detected plurality of character strings, the relationship information, and the distance.

4. The information processing device according to claim 1 or 2, wherein: The information processing device further comprises: a character recognition unit that performs character recognition based on the image; and correction department, The acquisition unit acquires correction information indicating a character string that is a result of the character recognition and a character string that is a result of the determination, which is a character string that is preferentially associated. The correction unit determines a character string to be associated based on the character string of the determination result, the character recognition result, and the correction information, and corrects the character string of the determination result to the determined character string when the character string of the determination result is different from the determined character string, that is, the determined character string.

5. The information processing device according to claim 1 or 2, wherein: The information processing device further comprises: a character recognition unit that performs character recognition based on the image; and correction department, The obtaining unit obtains correct answer information of the character string. The correction unit calculates the accuracy of the character recognition result, i.e., the character string of the character recognition result, based on the correct answer information, and calculates the accuracy of the character string of the determination result based on the correct answer information. When the accuracy of the character recognition result character string is greater than the accuracy of the determination result character string, the determination result character string is corrected to the character recognition result character string.

6. The information processing device according to claim 1, wherein: The information processing device further comprises a determination correction unit. The acquisition unit acquires correction information, which is information for correcting the determination result. The determination unit determines a combination of the first symbol and a character string closest to the first symbol based on the distance, and determines whether a relationship between the first symbol and the character string closest to the first symbol is registered in the relationship information. When the relationship indicated by the determination result is registered in the relationship information, the determination correction unit determines whether to correct the determination result based on the correction information, and when the determination result is to be corrected, corrects the determination result based on the correction information.

7. A processing method, wherein: The information processing device performs the following processing: acquiring an image including a first symbol and a plurality of character strings, and relationship information indicating a correspondence between the symbol and the character strings, extracting the first symbol and the plurality of character strings based on the image, and calculating a distance between the first symbol and the plurality of character strings based on a position of the first symbol and respective positions of the plurality of character strings, Based on the relationship information and the distance, a character string having a relationship with the first symbol is determined.

8. A processing procedure, wherein: The processing program causes the information processing device to execute the following processing: acquiring an image including a first symbol and a plurality of character strings, and relationship information indicating a correspondence between the symbol and the character strings, extracting the first symbol and the plurality of character strings based on the image, and calculating a distance between the first symbol and the plurality of character strings based on a position of the first symbol and respective positions of the plurality of character strings, Based on the relationship information and the distance, a character string having a relationship with the first symbol is determined.

Images