JP5655670B2 - Detection program, detection apparatus, and detection method - Google Patents

Description

  The present invention relates to a detection program, a detection apparatus, and a detection method for detecting a missing sentence.

  Conventionally, a screening facility has created a health check result report to notify the person of the check result of the health check, and has notified the person. In recent years, systematization of screening facilities has been attempted, and the system outputs a health check result report. However, since the system outputs the results of the health check entered by the attending physician within a limited area of the health check result report, if the number of characters to be output is large, it will be output with the sentence cut off halfway through. There was a case.

  As a technique for detecting the presence or absence of the output sentence, for example, after the printed form is recognized by an OCR (Optical Character Reader), it is searched whether the read sentence exists in the master. Thus, a technique for detecting missing text is disclosed (for example, see Patent Documents 1 and 2 below). As another technique for detecting missing sentences, for example, a comparison sentence is created from the target sentence, and the target sentence is created using a feature vector created from features such as the number of words and word types of the target sentence and the comparison sentence. Is disclosed to detect the lack of punctuation marks (see, for example, Patent Document 3 below).

JP-A-8-030700 Japanese Patent Laid-Open No. 2-158863 JP 2010-218318 A

  By applying the prior arts 1 to 3 described above, it is possible to detect that a sentence read by OCR is missing. However, the techniques according to the prior arts 1 to 3 have a problem that it takes time to detect a missing sentence in order to search for a sentence registered in the master or to acquire a feature. It was. For example, when there are a large number of health check result reports, the techniques according to the prior arts 1 to 3 have taken time to detect missing sentences.

  An object of the present invention is to provide a detection program, a detection apparatus, and a detection method capable of detecting a missing piece of text output in an area at high speed in order to solve the above-described problems caused by the prior art.

  In order to solve the above-described problems and achieve the object, according to one aspect of the present invention, a reading unit that reads characters from image data including an imaged sentence is present in a predetermined area of the image data. Next, the end character of the character string is detected, whether or not the end character is a punctuation point, and based on the result of the determination, the result of the possibility of missing subsequent sentences is determined. A detection program, a detection device, and a detection method for outputting are proposed.

  According to one aspect of the present invention, there is an effect that it is possible to detect a lack of text output in a region at high speed.

FIG. 1 is an explanatory diagram illustrating an operation example of the detection apparatus 100 according to the embodiment. FIG. 2 is a block diagram illustrating a hardware example of the detection apparatus 100. FIG. 3 is a block diagram illustrating an example of functions of the detection apparatus 100. FIG. 4 is an explanatory diagram illustrating an example of the health check result report 101. FIG. 5 is an explanatory diagram showing the contents stored in the ending check table 311. FIG. 6 is an explanatory diagram showing the contents stored in the character string field definition table 312. FIG. 7 is an explanatory diagram showing an example of the result of missing. FIG. 8 is a flowchart illustrating an example of missing sentence detection processing.

  Exemplary embodiments of a disclosed detection program, detection apparatus, and detection method will be described below in detail with reference to the accompanying drawings.

  FIG. 1 is an explanatory diagram illustrating an operation example of the detection apparatus 100 according to the embodiment. The detection apparatus 100 captures the printed health examination result report 101 using the scanner 102 and generates image data (hereinafter simply referred to as image data) 103 of the health examination result report 101.

  In addition, when the health check result report 101 is printed, among the target sentences “~~~~ suspected of cancer." This is a state in which the subsequent sentence “is out of the area 104 of the health check result report 101”. The user who viewed this report is healthy if the content of the report is an expression indicating the possibility of “suspected of cancer” or an expression indicating an assertion of “cancer” The result recognized as a result of diagnosis is completely different. The image data 103 includes the text “~ ˜˜˜cancer” existing in the region 104 in an imaged state. In the present embodiment, the printed health examination result report 101 is a detection target of a missing sentence, but a printed matter in which a file in which a sentence is input is printed may be a detection target.

  Subsequently, the detection apparatus 100 reads a character string existing in a predetermined area 104 from the image data 103 using the OCR. Hereinafter, a character string existing in the predetermined area 104 is referred to as a detection target character string. In the example of FIG. 1, the detection target character string is “˜˜˜˜cancer” existing in the region 104. After reading, the detection apparatus 100 detects the end character of the detection target character string, and subsequently determines whether the end character is a punctuation mark. In the example of FIG. 1, since the end character is “RU”, which is not a punctuation point, the detection apparatus 100 detects the lack of a subsequent sentence with respect to the imaged sentence.

  Thereby, since the detection apparatus 100 detects the presence or absence of a character string to be detected without searching a full-text registered database, it is possible to detect a subsequent sentence missing from an imaged sentence at high speed. Hereinafter, the hardware, function, and flowchart of the detection apparatus 100 that performs the operation illustrated in FIG. 1 will be described with reference to FIGS.

(Hardware of the detection apparatus 100)
FIG. 2 is a block diagram illustrating a hardware example of the detection apparatus 100. In FIG. 2, the detection apparatus 100 includes a CPU (Central Processing Unit) 201, a ROM (Read-Only Memory) 202, and a RAM (Random Access Memory) 203. The detection device 100 as a storage device includes a magnetic disk drive 204, a magnetic disk 205, an optical disk drive 206, and an optical disk 207. Further, the detection apparatus 100 as an input / output device for a user and other devices includes a display 208, an I / F (Interface) 209, a keyboard 210, a mouse 211, a scanner 102, and a printer 212. Each unit is connected by a bus 213.

  Here, the CPU 201 governs overall control of the detection apparatus 100. The ROM 202 stores a program such as a boot program. The RAM 203 is used as a work area for the CPU 201. The magnetic disk drive 204 controls reading / writing of data with respect to the magnetic disk 205 according to the control of the CPU 201. The magnetic disk 205 stores data written under the control of the magnetic disk drive 204.

  The optical disk drive 206 controls reading / writing of data with respect to the optical disk 207 according to the control of the CPU 201. The optical disk 207 stores data written under the control of the optical disk drive 206, or causes the computer to read data stored on the optical disk 207. Note that the detection program according to the present embodiment may be stored in any of the storage devices of the ROM 202, the magnetic disk 205, and the optical disk 207.

  The display 208 displays data such as a document, an image, and function information as well as a cursor, an icon, or a tool box. For example, the display 208 can employ a CRT, a TFT liquid crystal display, a plasma display, or the like.

  The I / F 209 is connected to a network 214 such as a LAN (Local Area Network), a WAN (Wide Area Network), and the Internet through a communication line, and is connected to other devices via the network 214. The I / F 209 controls an internal interface with the network 214 and controls data input / output from an external device. For example, a modem or a LAN adapter may be employed as the I / F 209.

  The keyboard 210 has keys for inputting characters, numbers, various instructions, and the like, and inputs data. The keyboard 210 may be a touch panel type input pad or a numeric keypad. The mouse 211 performs cursor movement, range selection, window movement, size change, and the like. In addition, the detection apparatus 100 may include a trackball, a joystick, or the like as long as it has a similar function as a pointing device instead of the mouse 211.

  The scanner 102 optically reads an image and takes in image data into the detection apparatus 100. The scanner 102 may have an OCR function. For example, the detection apparatus 100 according to the present embodiment reads a health check result report and reads a character string using the OCR function. The printer 212 prints image data and document data. As the printer 212, for example, a laser printer or an ink jet printer can be employed.

(Function of the detection apparatus 100)
Next, functions of the detection apparatus 100 will be described. FIG. 3 is a block diagram illustrating an example of functions of the detection apparatus 100. The detection apparatus 100 includes a reading unit 301, an acquisition unit 302, a detection unit 303, a determination unit 304, an output unit 305, and a search unit 306. The functions (reading unit 301 to search unit 306) serving as the control unit are realized by the CPU 201 executing the program stored in the storage device 310. Specifically, the storage device 310 is, for example, the ROM 202, the RAM 203, the magnetic disk 205, the optical disk 207, etc. shown in FIG. Alternatively, the function may be realized by another CPU executing via the I / F 209.

  Further, the detection apparatus 100 can access the ending check table 311, the character string field definition table 312, and the image data 103 included in the storage device 310. It is assumed that the image data 103 is data generated by reading the health check result report 101 by the scanner 102. Further, the image data 103 may be data acquired via the I / F 209.

  The ending check table 311 is a table in which a character string that can be set as the end character string of the detection target character string is stored. A specific data storage example of the ending check table 311 will be described later with reference to FIG. The character string field definition table 312 is a table that stores an area where the detection target character string in the health check result report 101 is printed. A specific storage example of the character string field definition table 312 will be described later with reference to FIG.

  The reading unit 301 has a function of reading characters from the image data 103 including the imaged text. For example, the reading unit 301 specifies a character by matching the image data 103 with a previously stored pattern, and reads the character. The function of the reading unit 301 may be in the scanner 102. The read character is stored in a writable storage area such as a register of the CPU 201 or the RAM 203.

  The acquisition unit 302 has a function of acquiring, from the reading unit 301, a detection target character string that is a detection target existing in a predetermined area of the image data 103. The predetermined area is an area defined in the character string field definition table 312. For example, the acquisition unit 302 acquires a detection target character string that exists in a rectangular area in which the upper left coordinates in the image data 103 are (10, 40) and the lower right coordinates are (300, 150). The acquired character string to be detected is stored in a writable storage area such as the register of the CPU 201 or the RAM 203.

  The detection unit 303 detects the last character of the detection target character string. For example, it is assumed that the detection target character string acquired by the acquisition unit 302 is “˜˜˜˜cancer”. At this time, the detection unit 303 detects “RU” as the last character. The detected end character is stored in a writable storage area such as the register of the CPU 201 or the RAM 203. The stored information may be a pointer indicating the position of the last character in the detection target character string stored in the acquisition unit 302.

  The determination unit 304 has a function of determining whether the last character is a punctuation mark. For example, it is assumed that the end character “ru” is detected by the detection unit 303. At this time, the determination unit 304 determines whether “RU” is a punctuation mark. The determination result is stored in a writable storage area such as a register of the CPU 201 or the RAM 203.

  The output unit 305 has a function of outputting a result of the possibility of missing subsequent text with respect to the imaged text based on the determination result by the determination unit 304. For example, the output unit 305 outputs that the subsequent sentence is missing when it is not a punctuation mark.

  Further, when the last character is not a punctuation, the output unit 305 may output that the subsequent sentence may be missing as a result of the possibility of the subsequent sentence being missing. Further, when the last character is a punctuation, the output unit 305 may output that the succeeding sentence is not missing as a result of the possibility of missing the succeeding sentence. In addition, as a result of the search by the search unit 306, the output unit 305 may output that the subsequent sentence is missing as a result of the possibility of the subsequent sentence being missing when there is no ending that matches the end character string.

  Further, the output unit 305 may output the result of the possibility of missing subsequent text and the end character based on the determination result by the determination unit 304. For example, the output unit 305 outputs, as a result of the possibility of missing subsequent text, that the subsequent text is missing and the end character is “RU”. The output format includes, for example, display on the display 208, print output to the printer 212, and transmission to an external device via the I / F 209. The output unit 305 may store the output result in a storage area such as the RAM 203, the magnetic disk 205, and the optical disk 207.

  When the determining unit 304 determines that the end character is not a punctuation point, the search unit 306 refers to the ending check table 311 in which the character string group related to the ending is stored, and the ending that matches the end character string of the detection target character string. It has a function to search.

  For example, a case is assumed where the detection target character string is “having cancer”. At this time, the determination unit 304 determines that the end character “ru” is not a punctuation point, and then the search unit 306 searches for a ending that matches the end character string of the detection target character string. For example, when there is a ending “Yes” in the ending check table 311, the search unit 306 outputs that the search result has been found, and otherwise outputs that the search result has not been found. The search result is stored in a writable storage area such as a register of the CPU 201 or the RAM 203.

  FIG. 4 is an explanatory diagram illustrating an example of the health check result report 101. In the health check result report 101, for example, a region 401 for displaying the name of the report, a region 402 for displaying the personal information of the subject of the health check, a region 403 for displaying the determination result, input by the doctor in charge An area 404 for displaying general findings is included. Further, the health diagnosis result report 101 includes an area 405 for displaying a message other than the general findings and an area 406 for displaying details of the diagnosis result. For example, the detection apparatus 100 stores an area 404 for displaying the general findings in the character string field definition table 312 in advance as the area 104 including the detection target character string.

  FIG. 5 is an explanatory diagram showing the contents stored in the ending check table 311. The ending check table 311 includes a field called ending. The ending field stores a character string that can be set as the last character string of the detection target character string. For example, the ending check table 311 shown in FIG. 5 stores “is”, “mass”, and “is”.

  FIG. 6 is an explanatory diagram showing the contents stored in the character string field definition table 312. The character string field definition table 312 includes three fields: file name, area name, and area. The file name field stores the name of the file that is the target for detecting the missing text. In the file name field, identification information for uniquely specifying the file may be stored. The area name field stores the name of the area containing the search target character string. The area field stores a specific range of area names. Specifically, the area field stores an upper left X coordinate, an upper left Y coordinate, a lower right X coordinate, and a lower right Y coordinate in the area.

  For example, in the character string field definition table 312 shown in FIG. 6, in the comprehensive findings area of the health check result report, the upper left coordinates are (10, 40) and the lower right coordinates are (300, 150). This indicates a rectangular area. In addition, as a setting method of an area field, it may be set by the system designer from, for example, data of a print source of a health check result report.

  FIG. 7 is an explanatory diagram showing an example of the result of missing. FIG. 7 shows four examples of missing results according to the detection target character string. The explanatory diagram denoted by reference numeral 701 indicates a missing case when the detection target character string shown in the area 104-1 of the health check result report 101-1 is “~ suspected to be cancer”. The result of the determination is shown. The explanatory diagram denoted by reference numeral 702 shows the result of determining omission when the detection target character string shown in the area 104-2 of the health check result report 101-2 is "~~~~ cancer". Show.

  In the explanatory diagram denoted by reference numeral 703, the character string to be detected shown in the area 104-3 of the health examination result report 101-3 is “to be cancer”, and the search by the ending check table 311 is further performed. The result of determining omission when added is shown. In the explanatory diagram denoted by reference numeral 704, the character string to be detected shown in the area 104-4 of the health check result report 101-4 is "~~~~ suspected to be cancer", and the ending check table 311 It shows the result of determining the lack when adding a search by.

  In the explanatory diagram denoted by reference numeral 701, since the last character of the detection target character string is “.” And is a punctuation mark, the detection apparatus 100 outputs that the subsequent sentence is not missing as a result of the possibility of missing. Further, the detection apparatus 100 may output the end character “.”. In the explanatory diagram denoted by reference numeral 702, since the end character of the detection target character string is “ru” and is not a punctuation mark, the detection apparatus 100 outputs that there is a possibility that the subsequent sentence may be missing as a result of the possibility of missing. Further, the detection apparatus 100 may output the last character “ru”. The reason why there is a possibility of missing in the state of reference numeral 702 is that there is a possibility that the sentence input to the area 104 is a sentence without a punctuation mark.

  In the explanatory diagram denoted by reference numeral 703, since the end character of the detection target character string is “ru” and is not a punctuation mark, the detection apparatus 100 continues to refer to the ending check table 311 to refer to the end character string of the detection target character string. Search for endings that match. At this time, assuming that the endings registered in the ending check table 311 are in the state shown in FIG. 5, the detection apparatus 100 determines that there is no ending that matches the end character string, and results in the possibility of missing. Output that the following sentence is missing. Further, the detection apparatus 100 may output the last character “ru”.

  In the explanatory diagram denoted by reference numeral 704, since the end character of the detection target character string is “su” and is not a punctuation mark, the detection apparatus 100 continues to refer to the ending check table 311 and ends the character string of the detection target character string. Search for endings that match. At this time, assuming that the ending registered in the ending check table 311 is in the state shown in FIG. 5, the detection apparatus 100 determines that there is a ending “Masu” that matches the end character string, and can be omitted. Outputs that the following sentence is not missing as a result of sex. Further, the detection apparatus 100 may output the end character “su”. In the state indicated by reference numeral 704, since there is a possibility of an input error, it may be output that there is a possibility of an input error.

  FIG. 8 is a flowchart illustrating an example of missing sentence detection processing. The detection apparatus 100 selects the first health examination result report 101 from the group of health examination result reports 101 (step S801). After the selection, the detection apparatus 100 captures the selected health check result report 101 using the scanner 102 and generates image data 103 (step S802). After generation, the detection apparatus 100 acquires, from the image data 103, the detection target character string in the region field of the character string field definition table 312 using OCR (step S803).

  After acquisition, the detection apparatus 100 detects the end character of the detection target character string (step S804). After detection, the detection apparatus 100 determines whether the last character is a punctuation mark (step S805). When the last character is a punctuation (step S805: Yes), the detection apparatus 100 outputs that the subsequent sentence for the imaged sentence is not missing (step S806).

  When the last character is not a punctuation (step S805: No), the detection apparatus 100 refers to the ending check table 311 and searches for a ending that matches the last character string of the detection target character string (step S807). After the search, the detection apparatus 100 determines whether a matching ending is found (step S808). If not found (step S808: No), the detection apparatus 100 outputs that the subsequent sentence for the imaged sentence is missing (step S809). If found (step S808: Yes), the detection apparatus 100 outputs that there is a possibility of an input error in the subsequent sentence with respect to the imaged sentence (step S810).

  After outputting in step S806, step S809, and step S810, the detection apparatus 100 determines whether or not all the health examination result reports 101 have been selected (step S811). When all the health examination result reports 101 are not selected (step S811: No), the detection apparatus 100 selects the next health examination result report 101 (step S812), and proceeds to the process of step S802. When all the health examination result reports 101 have been selected (step S811: Yes), the detection apparatus 100 ends the missing sentence detection process.

  In step S806, the detection apparatus 100 may not perform the output performed in step S806, assuming that the detection result is normal. If there is a large amount of health check result report 101 of missing sentence detection when outputting, the output destination log is filled with information that “not missing”. As a result, it may be difficult for the log checker to find the “missing” health check result report 101. By not performing the output, it becomes easy for the log checker to find the “missing” health check result report 101.

  In addition, in step S805: No route, the detection apparatus 100 does not perform the processing in steps S807 and S808, but outputs that “possibly missing” in step S809: No route, You may transfer to the process of S811. Thereby, the detection apparatus 100 can detect that the subsequent sentence with respect to the sentence imaged with less processing is missing.

  Note that, in the output processing performed in the processing of step S806, step S809, and step S810, the detection apparatus 100 determines whether the subsequent sentence for the imaged sentence is missing and the end of the detection target character string. Characters may be output.

  As described above, according to the detection program, the detection device, and the detection method, the character string read from the imaged report in the process of detecting the lack of the subsequent sentence that protrudes from the print area of the printed file. Whether the following sentence is missing is determined by whether or not the end of the sentence is a punctuation mark. As a result, the detecting device does not need to search a database in which the full text printed on the report is registered in order to detect the missing text, so that the process for determining the presence or absence of the missing text can be performed at high speed. Conventionally, for example, the screening staff has confirmed visually whether there is a contradiction in the outputted health check result report. Thus, by automating the confirmation work, it is possible to reduce the burden on the screening staff.

  In addition, when the last character is not a punctuation point, the detection device may output that the subsequent sentence may be missing as a result of the possibility that the subsequent sentence is missing from the imaged sentence. Thereby, the detection apparatus can detect the text which protruded from the printing area | region in a form. In addition, when the last character is a punctuation, the detection device may output that the subsequent sentence is not missing as a result of the possibility of missing the subsequent sentence with respect to the imaged sentence. Thereby, since the screening staff does not need to confirm the output form when the subsequent text is not missing, the burden on the screening staff can be reduced.

  In addition, when the last character is not a punctuation, the detection device refers to the table storing the character string group related to the ending and searches for the ending that matches the last character string of the detection target character string, and matches the last character string. If there is no ending, it may be output that the subsequent sentence is missing as a result. As a result, the detection device detects that there is a possibility of an input error because the input sentence is incorrect when the last character is not a punctuation but the sentence that was originally input has no punctuation. Can do.

  Further, the detection device may output a result of the possibility of missing a subsequent sentence and an end character with respect to the imaged sentence. Thereby, the screening staff can confirm the end character by confirming the output result without confirming the health examination result report.

  In addition, the detection program according to the present embodiment has explained the target to detect the missing text, taking a health check result report as an example, but it can also be used as a check program for forms with limited output areas. You can also

  The detection method described in this embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. The detection program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk, a CD-ROM, an MO, and a DVD, and is executed by being read from the recording medium by the computer. The detection program may be distributed through a network such as the Internet.

  The following additional notes are disclosed with respect to the embodiment described above.

(Supplementary note 1)
From a reading unit that reads characters from image data containing an imaged sentence, a character string existing in a predetermined region of the image data is acquired,
Detect the last character of the string;
Determine whether the last character is a punctuation mark,
Based on the result of the determination, the result of the possibility of missing subsequent sentences for the sentence is output.
Detection program that executes processing.

(Supplementary note 2) The output process is
If the last character is not the punctuation, output that the subsequent sentence may be missing as a result.
The detection program according to supplementary note 1, characterized in that:

(Supplementary note 3)
If the end character is the punctuation, the result is that the subsequent sentence is not missing,
The detection program according to supplementary note 1, characterized in that:

(Supplementary Note 4) When the end character is not the punctuation point, the end character string that matches the end character string of the character string is searched with reference to the table storing the character string group related to the end character.
The output process is as follows:
If there is no ending that matches the end string, the result is that the subsequent sentence is missing,
The detection program according to supplementary note 2, characterized by:

(Supplementary note 5)
Based on the result of the determination, the result and the end character are output.
The detection program according to any one of supplementary notes 1 to 4, wherein:

(Additional remark 6) The acquisition part which acquires the character string which exists in the predetermined area | region of the said image data from the reading part which reads a character from the image data containing the imaged sentence,
A detection unit for detecting a last character of the character string;
A determination unit for determining whether the last character is a punctuation;
Based on the result of the determination by the determination unit, an output unit that outputs a result of the possibility of missing subsequent sentences for the sentence;
A detection apparatus comprising:

(Appendix 7) The computer
From a reading unit that reads characters from image data containing an imaged sentence, a character string existing in a predetermined region of the image data is acquired,
Detect the last character of the string;
Determine whether the last character is a punctuation mark,
Based on the result of the determination, the result of the possibility of missing subsequent sentences for the sentence is output.
A detection method that performs processing.

DESCRIPTION OF SYMBOLS 100 Detection apparatus 101 Health check result report 102 Scanner 103 Image data 201 CPU
213 Bus 301 Reading unit 302 Acquisition unit 303 Detection unit 304 Determination unit 305 Output unit 306 Search unit 310 Storage device 311 End check table 312 Character string field definition table

Claims (5)

On the computer,
From a reading unit that reads characters from image data containing an imaged sentence, a character string existing in a predetermined region of the image data is acquired,
Detect the last character of the string;
Determine whether the last character is a punctuation mark,
If the last character is not the punctuation, output that the subsequent sentence may be missing as a result of the possibility of missing the subsequent sentence for the sentence ;
Detection program that executes processing. The output process is as follows:
  If the end character is the punctuation, the result is that the subsequent sentence is not missing,
The detection program according to claim 1. In the computer,
If the end character is not the punctuation mark, refer to a table storing character string groups related to the end of the word, search for the end of the character string that matches the end character string of the character string,
  The output process is as follows:
If there is no ending that matches the end string, the following sentence is missing as a result.
Output that
The detection program according to claim 1. An acquisition unit that acquires a character string existing in a predetermined region of the image data from a reading unit that reads characters from image data including an imaged sentence;
A detection unit for detecting a last character of the character string;
A determination unit for determining whether the last character is a punctuation;
If the end character is not the punctuation point, an output unit that outputs that the subsequent sentence may be missing as a result of the possibility of missing the subsequent sentence for the sentence; and
A detection apparatus comprising: Computer
From a reading unit that reads characters from image data containing an imaged sentence, a character string existing in a predetermined region of the image data is acquired,
Detect the last character of the string;
Determine whether the last character is a punctuation mark,
If the last character is not the punctuation, output that the subsequent sentence may be missing as a result of the possibility of missing the subsequent sentence for the sentence;
A detection method that performs processing.

Images