JP2016071898A - Form recognition device, form recognition system, program for form recognition system, control method for form recognition system and recording medium loaded with form recognition system program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a form recognition device, a system, a method and a form recognition program capable of accurately recognizing letters in a form.SOLUTION: A form recognition device includes: a region determination unit for recognizing a document of a constant format including a form and determining a region regarding line or frame information; a character within region recognition unit for recognizing a character within the region for a specified region within a determined region; an item corresponding to attribute selection unit for selecting an item corresponding to an attribute estimated to be item name information among character information obtained by the character within region recognition means from an attribute-possible item within attribute data base; an adjacent item selection unit for selecting one with high degree of similarity regarding character information other than regarding the information estimated to be the item name information; and a character information confirmation unit for verifying and confirming adequacy of the character information selected by the adjacent item selection means. Furthermore, a plurality of formula expression patterns are associated with a same form pattern. In addition, the displayed digital information is corrected by displaying the recognized digital information.SELECTED DRAWING: Figure 11

Description

  The present invention relates to form recognition, and in particular, a form recognition apparatus, a form recognition system, a form recognition system program, a form recognition system control method, and a form recognition system program that can automatically recognize prescriptions in dentistry and medical departments. The present invention relates to a recording medium equipped with.

  In the field of dentistry and medical treatment related to forms, especially prescriptions, doctors issue prescriptions for necessary prescriptions, bring them to pharmacies, and prescribe necessary medicines. Currently, in general pharmacies, prescriptions brought in are visually recognized by pharmacy people and then manually entered into the system or processed by limited recognition technology.

  In addition, doctors may rush to write while examining the patient, and many prescriptions are prescriptions. Therefore, it cannot be denied that the above-mentioned limited recognition technique has many misrecognitions. Therefore, in some cases, a prescription is printed by a printer or the like by applying an electronic device such as a personal computer.

  On the other hand, as a matter of course, when a misread medicine is prescribed, there is a possibility that the medicine may be important to human life.

  For example, in Patent Document 1, in the prescription receiving apparatus 10 that takes a prescription by the prescription scanning unit 20 and obtains prescription data from the image image by the data processing unit 30, if the image image has a two-dimensional code, Is converted to prescription data, and when there is no two-dimensional code, character recognition is performed on the image image to convert it to prescription data. At that time, the medical institution is specified and the prescription format is selected and the image image is divided In addition, a technical idea is disclosed in which the prescription data can be input efficiently by further displaying the divided image and the conversion field in comparison and the divided image and the calibration field.

  However, in patent document 1, since it is limited to a standard prescription format, it cannot respond to any prescription format. In addition, since the recognition of the characters in the prescription is not different from the previous one, it has not been realized to eliminate the misrecognition.

  Some conventional reading systems for forms, particularly prescriptions, take a lot of man-hours for the prior learning work when the system is introduced. This is because, for each prescription template, it is necessary to designate a field to be character-recognized by human input and to let the software learn the meaning of the output result of each field. In addition, the character recognition process for each designated recognition target field also requires teacher learning by a person to correct and replace the recognition result of OCR (abbreviated notation of the optical character reader) each time. This work requires learning thousands of prescriptions for each medical institution in order to increase OCR recognition accuracy. About these learning, the same learning is necessary with the prescription layout change after introduction, and the maintenance cost is also borne.

  In addition to this, in the above-mentioned current system, it is necessary for the pharmacy to provide the original prescription that is actually used to learn the prescription. It must be said that the specifications are in conflict with the request to finish the work. For this reason, the form of the prescription that can be recognized is limited by the balance of the man-hours required for template registration and OCR learning work, and this is an obstacle to widespread use.

  Patent Document 2 discloses various drug names in the prescription accepting apparatus 10 that takes a prescription by the prescription scanning unit 20 and applies prescription data from the image image by applying a technique such as OCR from the image image. A technical idea is disclosed in which a medicine master including a prescription data obtained from an image is compared and the result is reflected in a character string replacement table.

  However, Patent Document 2 discloses a technical idea limited to correcting an error with respect to a specific character string, and does not disclose a technical idea of eliminating erroneous recognition of the drug name itself. The technical idea of accurately recognizing the character string described in the prescription is not reached.

JP 2006-120020 A JP 2007-164274 A

  The present invention provides a prescription recognition apparatus, system, program, method, and recording medium equipped with a prescription recognition program capable of accurately recognizing a form, particularly a prescription character, based on the above problems. Is an issue.

More specifically, the present invention will be described by taking a form, particularly a prescription as an example.
・ I can improve the character recognition rate in the `` prescription column '' in the prescription,
・ Prescription format registration / change work can be simplified.
・ Read prescriptions printed on various printers.
As described above, the prescription has been described as an example, but the present invention enables application to all forms other than the prescription, for example, business transactions, official documents,
It is an object to provide a form recognition apparatus, a form recognition system, a form recognition system program, a form recognition system method, and a recording medium equipped with a form recognition program.

  In order to solve such a problem, the form recognition apparatus according to the present invention recognizes ruled lines and / or frame information of a certain type of document including the form, and determines an area related to the recognized ruled lines and / or frame information. For a specific area among the areas determined by the area determination unit and the area determination means, the characters in the area are imaged and reduced to digital information, and the character information that matches the reduced digital information is digitally converted. An in-region character recognition unit that obtains corresponding character information by extracting from the information-character information correspondence database; Alternatively, the estimation and / or the determination from the attribute-attribute-possible item database in which an attribute and a list of item information that can be assumed to belong to the attribute are associated with each other. Attribute corresponding to the specified attribute-Attribute corresponding item selection unit for selecting a possible item database within the attribute, and characters other than those related to the information estimated as item name information among the character information obtained by the in-area character recognition means For the information, the item selected by the attribute corresponding item selection unit is selected by the proximity item selection unit that selects the item information stored in the attribute-in-attribute possible item database with a high degree of similarity and the proximity item selection unit A character information confirmation unit that verifies the validity of the item information with high similarity and confirms information corresponding to the character information, and further includes the region determination unit and the in-region character recognition unit. A plurality of prescriptions for the same form pattern with respect to the form-specific format recognized by the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit. And means for associating a serial pattern, and means for storing the information about the result of the associating a first memory. In addition, regarding the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit, ruled lines, handwritten characters, and print character information are digital information Means for displaying at least one of the recognized results, means for correcting the displayed digital information, and means for storing information relating to the corrected result as a second storage.

  In order to solve the above problems, a form recognition system according to the present invention is recognized by a frame line recognition apparatus that recognizes ruled lines and / or frame information of a document in a certain format including a form, and the frame line recognition apparatus. Area determining means for determining an area related to the ruled line and / or frame information, and for a specific area among the areas determined by the area determining means, imaging characters in the area and reducing it to digital information, Character information matching the reduced digital information is extracted from the digital information-character information correspondence database to obtain the corresponding character information, and the item name of the character information obtained by the in-region character recognition means Attributes that presume and / or specify attributes related to information and estimated information, and associate attributes with a list of item information that can be assumed to belong to the attribute Attribute corresponding to the estimated and / or specified attribute from the attribute-possible item database-attribute-corresponding item selection means for selecting the attribute-possible item database, and item name of the character information obtained by the in-area character recognition means Proximity item that selects the item information stored in the attribute-attribute-possible item database selected by the attribute-corresponding item selection means for character information other than information that is estimated to be information. Selecting means, and character information determining means for verifying the validity of the item information with high similarity selected by the proximity item selecting means and determining information corresponding to the character information, Recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit And relates format of the form, and means for associating a plurality of prescription notation pattern for the same form patterns, and means for storing the information about the result of the associating a first memory. In addition, regarding the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit, ruled lines, handwritten characters, and print character information are digital information Means for displaying at least one of the recognized results, means for correcting the displayed digital information, and means for storing information relating to the corrected result as a second storage.

  Furthermore, in order to solve the above-described problem, the form recognition method according to the present invention recognizes ruled line and / or frame information of a document in a certain format including the form, and an area related to the recognized ruled line and / or frame information. A region determining step for determining the character, and for a specific region of the determined region, the characters in the region are imaged and reduced to digital information, and the character information that matches the reduced digital information is converted into digital information. -Intra-region character recognition step for obtaining corresponding character information by extracting from the character information correspondence database; estimating and / or defining attributes relating to information estimated as item name information among the obtained character information; Attribute corresponding to a list of item information that can be assumed to belong to the attribute-attribute corresponding to the attribute estimated and / or specified from the attribute-possible item database -The attribute correspondence item selection step for selecting the attribute-possible item database and the character information other than the item name information and the information estimated from the obtained character information were selected by the attribute correspondence item selection means. Attribute-neighbor item selection step for selecting item information having a high similarity among item information stored in the attribute-possible item database, and verifying the validity of the selected item information having the high similarity. A character information determination step for determining information corresponding to the information, and further comprising the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination. The step of associating a plurality of prescription patterns with the same form pattern, and the result of the association And a step of storing information as the first storage. In addition, regarding the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit, ruled lines, handwritten characters, and print character information are digital information And displaying at least one of the recognized results, correcting the displayed digital information, and storing information on the corrected result as a second storage.

  Furthermore, in order to solve the above-described problem, a form recognition program according to the present invention recognizes ruled lines and / or frame information of a certain type of document including a form, and recognizes the recognized ruled lines and / or frames. An area determination unit for determining an area related to information, and for a specific area among the areas determined by the area determination unit, characters in the area are imaged and reduced to digital information, and the reduced digital information The character information that matches the character information is extracted from the digital information-character information correspondence database to obtain the corresponding character information, and the information estimated as the item name information among the character information obtained by the character recognition means Attribute that can be estimated and / or specified and associated with a list of item information that can be assumed to belong to the attribute The attribute corresponding item selection means for selecting the attribute-attribute-possible item database corresponding to the estimated and / or specified attribute from the database, and the item name information is estimated among the character information obtained by the in-region character recognition means. For the character information other than the information related to the attribute, the item selected by the attribute corresponding item selection means-the proximity item selection means for selecting the item information having a high similarity among the item information stored in the in-attribute possible item database; A character information determination unit that verifies the validity of the item information having a high similarity selected by the proximity item selection unit and determines information corresponding to the character information; Recognized by the determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit Relates format votes, and means for associating a plurality of representation pattern for the same form pattern, means for storing the information about the result of the associating a first memory, that operates as a. In addition, regarding the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit, ruled lines, handwritten characters, and print character information are digital information And a means for displaying at least one of the recognized results, a means for correcting the displayed digital information, and a means for storing information relating to the corrected result as a second storage. However, the “frame” here is not only a closed area surrounded by ruled lines, but also a part of the area where some ruled lines are missing or not surrounded by ruled lines at all if it meets the requirements. It shall be recognized as “frame”. Furthermore, “form” refers to a document for information notation typified by “prescription”, and “form pattern” is an outline format for information notation typified by “prescription pattern”. Shall be pointed to.

  In the form recognition device, form recognition system, form recognition system program, form recognition system method, and recording medium equipped with the form recognition program according to the present invention, a specific area is recognized based on the ruled line of the form and thus recognized. Since the title (for example, the description name in the description column) is recognized for the characters in the specified area, the item name that has been character-recognized can be identified from the list of item names that can best match the recognized description name. Thus, it is possible to reduce the number of erroneous recognition in the character recognition of the form.

More specifically, according to the present invention,
・ Simplification of template registration work is achieved by automatic recognition of form templates.
・ Improves the recognition accuracy of the product name and usage listed in the product name column, and can greatly reduce the man-hours required for OCR learning work.
-By using the image processing technology according to the present invention, it is possible to correspond to a form format that cannot be recognized by the current system.
・ Enables recognition of forms printed on various printers.
This is an effect unique to the present application.

It is a functional block diagram of the whole form recognition system which concerns on one Embodiment of this invention. It is a block diagram of the management system which concerns on one Embodiment of this invention. It is a lineblock diagram of a branch office system concerning one embodiment of the present invention. It is an example figure of a prescription concerning one embodiment of the present invention. It is a ruled line information data table concerning one embodiment of the present invention. It is a prescription closed area information data table concerning one embodiment of the present invention. It is a prescription pattern data table concerning one embodiment of the present invention. It is a prescription reading information data table concerning one embodiment of the present invention. It is the prescription detail which concerns on one Embodiment of this invention. It is a prescription description pattern list concerning one embodiment of the present invention. It is a prescription recognition system process outline figure concerning one embodiment of the present invention. It is a horizontal line part example figure which concerns on one Embodiment of this invention. It is a pixel value acquisition schematic according to an embodiment of the present invention. It is an angle median value processing schematic diagram concerning one embodiment of the present invention. It is a distributed processing 1 schematic diagram concerning one embodiment of the present invention. It is a distributed processing 2 schematic diagram concerning one embodiment of the present invention. It is a horizontal-vertical pixel total schematic diagram concerning one Embodiment of this invention. It is a foam comparison schematic diagram concerning one embodiment of the present invention. It is a division area adjustment schematic diagram concerning one embodiment of the present invention.

  Hereinafter, with reference to the drawings, a form recognition apparatus, a form recognition system, a form recognition system program, a form recognition system method, and an outline of a recording medium equipped with a form recognition program according to an embodiment of the present invention will be described. Will be described as an example. In the following, the range necessary for the description for achieving the object of the present invention is schematically shown, and the range necessary for the description of the relevant part of the present invention will be mainly described. According to a known technique.

The present application performs prescription recognition according to the following procedure.
1. Prescription paper registration a. Automatic setting of entry frame i. Recognize ruled lines from scanned blank / filled prescription forms.
ii. The inclination is corrected using ruled line information.
iii. Character recognition is performed on the image from which the ruled lines have been removed, and the label character string of the entry frame such as “name” and “insurer number” is extracted.
iv. The position and type of the entry frame are estimated from the ruled line information and the position of the label character string. Anything that could not be estimated or made a mistake will be corrected by the user.
b. Extraction of registration features i. As a feature for aligning the prescription paper, calculate a horizontal / vertical histogram of the prescription paper in the unfilled state (if prefilled prescription paper is used for registration, based on the information set in a, (Calculate after automatically blanking out the entry frame).
c. Extraction of candidate refinement features i. Features that can be easily calculated and compared, such as the aspect ratio of the prescription paper area, histogram, and grid points, are calculated as features for narrowing down candidates.
d. Extracting matching features i. The prescription paper area in the image is subdivided into rectangles, and only the part that does not match the entry frame is taken out.
ii. Features such as color, histogram, average, variance, etc. are calculated for each rectangular portion extracted at i.

2. Prescription paper recognition a. Similar to prescription paper registration, it recognizes ruled lines and performs tilt correction.
b. Similar to the prescription paper registration, the registration feature and the candidate narrowing feature are calculated.
c. For each registered prescription paper, a comparison is made based on candidate narrowing characteristics, and only those whose differences do not exceed a certain threshold are extracted as candidates.
d. c. For each form of the result of the above, the alignment by the alignment feature and the comparison by the matching feature are performed, and those whose differences do not exceed a certain threshold are output as recognition results.

3. Prescription content recognition a. Correction of OCR recognition result i. Characters that are easily misrecognized by OCR are classified as one class, thereby reducing the influence of misrecognition on similar characters.
ii. Correction of misrecognized characters by trigram (For prescription contents, it can be seen that only the word “root” comes between “Kudzu” and “Yu”. , Can be limited to several patterns).
b. Recognizing Drug Names When recognizing drug names, a drug name dictionary is used to recognize and recognize drug names by using an algorithm that searches for and determines a character string that is highly similar to a drug in the dictionary. The rate is improving.
c. Quantity recognition i. From the OCR recognition result, the part of [numerical value] [unit] is extracted as a quantity candidate.
ii. b. The quantity candidate closest to the drug recognized in step 3 is set as the quantity of the drug.
d. Usage recognition i. The words and phrases used in usage are ranked in terms of frequency of use and importance, and points are assigned to the words and phrases.
ii. i. Based on the score set in step 1, a score is assigned to each line of the OCR recognition result.
iii. A line whose score exceeds a certain threshold is defined as a usage fragment.
iv. If the score of a line adjacent to the usage fragment is positive, that line is also used. However, this does not include cases where there is a straight line between the usage fragment and the line.
v. When the usage fragment can no longer be expanded, the usage fragment is used.
However, parts that are judged as drug names, quantities, etc. are excluded.

  An outline of a prescription recognition system according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a functional block diagram showing an outline of the configuration of a prescription recognition system according to an embodiment of the present application. A prescription recognition system according to an embodiment of the present invention is configured by connecting a management system 20 and a sales office system 30 via a network 10. In addition, it may be connected to the related institution system 202 or the like via the network 10 as necessary, and the system configuration may be configured by so-called cloud computing as necessary. Further, for example, necessary software or the like may be installed from the management system 20 to the sales office system 30. The related institution system 202 includes, for example, each system related to the prescription recognition system according to the present application, such as a database system related to the managing director of the Ministry of Health, Labor and Welfare, various prefectural medical related systems, and a database system related to drug names by the Japanese Pharmacopoeia. And

  Next, details of the management system 20 will be described with reference to FIG. As shown in FIG. 1, the management system 20 includes a management device database (hereinafter referred to as “management device DB”) 21, a management device server 22, a management output device 23, a management input device 24, and a management device interface 26. It is configured. The management apparatus server 22 processes data related to management of the prescription recognition system by a processor or the like, and the management apparatus DB 21 stores data such as management and information on the prescription recognition system. The management output device 23 includes a display, a printer, and the like, and outputs various types of information as necessary. The management input device 24 includes a keyboard, a barcode reader, a scanner, and the like, and inputs information as necessary. However, the management input device 24 includes all devices that can input information. The management system 20 may be composed of a single system or a plurality of systems as long as the management system 20 can finally perform the business as a prescription recognition system.

  Next, details of the sales office system 30 will be described with reference to FIG. As shown in the figure, a branch office system 30 that specifically performs a prescription reading operation such as a pharmacy, for example, has a branch store device memory 31, a branch store processor 32, a branch store output device 33, a branch store input device 34, A branch office device interface 36 is provided. The sales office device processor 32 processes, for example, information sent from the management system 20 and displays the information on the sales office output device 33, for example. The store output device 33 includes a display, a printer, and the like, and outputs various types of information as necessary. The sales office input device 34 includes a keyboard, a barcode reader, a scanner, and the like, and an operator inputs information as necessary, but includes all devices that allow information to be input. . In the store input device 34, input processing is performed from a keyboard or the like as necessary, for example, displayed on the store output device 33, and if necessary, information input from the store input device 34 by the operator. Then, the data is transmitted to the management system 20 via the sales office device interface 36. The management system 20 performs management processing of the prescription recognition system in accordance with the transmitted information, and when further information is required, further information input is performed by the operator from the sales office system 30. Thus, the operator operates the sales office system 30 and reads the prescription by the prescription recognition system.

  The outline and configuration of an embodiment of the present invention have been described above. Note that each system other than the management system 20 and the sales office system 30 also has a processor, a memory, etc. (all not shown), as in the management system 20 and the sales office system 30, as necessary. It is assumed that various information processing is performed. Next, details and operations of the present invention will be described.

  First, an example of a prescription read in the present application is shown in FIG. The prescription is created in accordance with the Ordinance of the Ministry of Health, Labor and Welfare, etc., and FIG. 40 is an example of the prescription. Prescription Example In the upper part of FIG. 40, a notation “prescription (or prescription)” indicating that it is a prescription is clearly written in the prescription column 401, and a prescription number column 41 indicating a public payer number, etc., and patient information are shown. It includes a patient information column 42, a prescription column 43 indicating specific prescription contents, a remarks column 44 indicating remarks, a dispensing related column 45 indicating information related to dispensing, a medical institution information column 46 indicating medical institution information, and the like. Prescriptions are created in advance by printing, etc., in the medical institutions that issue prescriptions, and the required items are filled in, or the prescriptions are filled in on the prescription paper, etc. In addition to printing the input information as a prescription, the input information is generated by a process such as printing as a form including mandatory items including ruled lines. Although these prescriptions are prescribed for required items, etc., they do not have a predetermined format. For example, companies, corporations, or medical associations that build systems that issue prescriptions, Various styles may exist depending on the designation of local governments. In the present application, the goal is to deal with any prescription style, and if available, a sample or the like according to the prescription style may be obtained and read in advance. In addition, the character string of the prescription is read and identified by, for example, OCR, and the position of the prescription notation is located or the direction of the character itself is identified, thereby identifying the head portion, the top and the like of the prescription. It doesn't matter.

  Next, for example, a method of automatically recognizing a “closed area” that is an area surrounded by a ruled line included in a form by, for example, a format analysis system (FGS) is shown in FIG. The ruled line information data table 50 that stores information related to ruled lines in the read prescription will be described first. The ruled line information data table 50 includes a ruled line information management column 51 for managing the ruled line data table, a ruled line 1 information column 5201, a ruled line 2 information column 5202, a ruled line 3 information column 5203,. I have. Further, the ruled line information management column 51 includes a number column 511 for sequentially numbering 1, 2, 3, and the like, a start end column 5101, an XY column 5102, and a start column 51011 includes a start X column. 512, the start Y column 513, the end column 51012 are configured to further include an end X column 514, an end Y column 515,. The ruled line information data table 50 is, for example, information calculated from related information by the branch store processor 32 according to a program stored in the branch store memory 31 or ruled line information displayed on the display of the branch store output device 33. In accordance with a data table setting screen (not shown), for example, information input by an operator from the keyboard of the sales office input device 34 of the sales office system 30 according to a program stored in the sales office device memory 31, the sales office device processor It is assumed that the data is stored on the store apparatus memory 31 by 31. The ruled line information data table setting screen (not shown) has the same format as the ruled line information data table 50. Further, if necessary, a ruled line information data table display screen (not shown) according to the contents of the ruled line information data table 50 is displayed on, for example, the store output device 33 of the store system 30 and the management output device 23 of the management system 20. Shall be displayed. The ruled line information data table display screen (not shown) has the same format as the ruled line information data table 50. The ruled line information data table 50 may be created and / or input on the management system 20 side and stored in the memory. Further, the information input and stored on the sales office system 30 side may be transmitted to the management system 20 side via the network 10, for example, and created and / or input on the management system 20 side and stored in the memory. Information may be transmitted to the branch office system 30 side.

  Regarding the ruled line information data table 50 configured as described above, in the ruled line information management column 51, for example, the number column 511 has “number”, the start column 51011 has “start”, the end column 51012 has “end”, etc. For example, “X” is stored in the start X column 512, “Y” is stored in the start Y column 513, “X” is stored in the end X column 514, and “Y” is stored in the end Y column 515, for example. Note that the X axis is, for example, the horizontal direction when viewing the prescription example FIG. 40, and the vertical direction is the Y axis. Next, in the number column 511 such as the ruled line 1 information column 5201, the ruled line 2 information column 5202, the ruled line 3 information column 5203,..., For example, “1”, “2”, “3”,. Is stored. Next, the data is stored in the start X column 512, the start Y column 513, the end X column 514, the end Y column 515, etc., such as the ruled line 1 information column 5201, the ruled line 2 information column 5202, the ruled line 3 information column 5203,. The calculation method of each information is shown below. First, a prescription is read by, for example, a scanner (not shown) connected to a store input device 34 controlled by the store device processor 32 or the like according to a program developed on the store system 30, for example. (Not shown) is stored in, for example, the branch office device memory 31. Next, the closed area is automatically recognized from the ruled lines and / or frames (hereinafter simply referred to as ruled lines) of the read prescription, and the positional relationship of each closed area and the character string in the closed area are, for example, as described below. By OCR recognition, which column of the prescription each closed region corresponds to is automatically extracted, and the inclination is corrected as described later from the information obtained by reading the prescription. Next, based on the read prescription information, if the prescription format is, for example, according to the prescription example FIG. 40, for example, the detection processing of the ruled line drawn first from the ruled line end 411 to the ruled line end 4198 with respect to the ruled line in the X-axis direction. Further, in the detected ruled line, with respect to the ruled line end 411 to the ruled line end 412, for example, the coordinate in the X direction indicating the position of the ruled line end 411 is, for example, 4.229 in millimeter units (the same applies hereinafter) If the coordinate is, for example, 15.857, the coordinate in the X direction indicating the position of the ruled line end 412 is, for example, 23.257, and the coordinate in the Y direction is, for example, 15.857, the start X in the ruled line 1 information column 5201 For example, “4.229” in the column 512, “15.857” in the start Y column 513, “23.257” in the end X column 514, and “15.257” in the end Y column 515, for example. The information that the 57 "is stored, respectively. Similarly, the detection processing is performed on the ruled line drawn from the ruled line end 413 to the ruled line end 414, and for example, “4.229” in the start X column 512 and “24.314” in the start Y column 513 of the ruled line 2 information column 5202. In the end X column 514, for example, information “23.257” is stored in the end Y column 515, for example “24.314”. Further, regarding the ruled line in the Y-axis direction, a detection process is performed on the ruled line drawn from the ruled line end 411 to the ruled line end 413, and for example, “4.229” is entered in the start X column 512 of the ruled line 3 information column 5203. For example, information such as “15.857” is stored in 513, “4.229” is stored in the end X column 514, and “24.314” is stored in the end Y column 515, for example. Similarly, for the ruled line end 413 to the ruled line end 414, for example, “23.257” in the start X column 512, “15.857” in the start Y column 513, and “15.857” in the end X column 514 in the ruled line 4 information column 5204. 23.257 ”and the information“ 24.314 ”is stored in the end Y column 515, for example. However, if no ruled line is detected, for example, the start X field 512 and the start Y field 513 of the ruled line 1 information field 5201, and the end X field 514 and the end Y field 515 of the end field 51012 are all For example, information “0” is stored to indicate that a ruled line could not be detected. The inclination correction when the ruled line cannot be detected may be performed by detecting a character string of “prescription (or prescription)” and correcting from the inclination of the character string. Regarding the detection of the ruled line, for example, when a prescription is read by a scanner and it is difficult to detect it because it is read with an inclination exceeding an allowable range, the process of reading from the scanner may be performed again. . The various detection processes and the like are processed by the branch store processor 32 according to a program developed on the branch store system 30, for example. The same processing may be performed by the management system 20. Further, the result may be transmitted to the sales office system 30 via the network 10, for example. The same processing may be performed by the management system 20. Further, the result may be transmitted to the sales office system 30 via the network 10, for example.

  For example, information on the ruled line 1 information column 5201, the ruled line 2 information column 5202, the ruled line 3 information column 5203, and the ruled line 3 information column 5203 stored in the ruled line information data table 50 includes the ruled line end 411 of FIG. A rectangular shape defined by an end 412, a ruled line end 413, and a ruled line end 414 is shown. Here, for example, a neighborhood point of the ruled line end 411 is set, and, for example, a straight line extending in all directions from the neighborhood point has a rectangular shape defined by the ruled line end 411, the ruled line end 412, the ruled line end 413, and the ruled line end 414. When two ruled lines intersect simultaneously at one of the two neighboring points, the neighboring point does not exist in the defined rectangular shape. On the other hand, when the straight line always intersects with a rectangular ruled line at two points sandwiching the neighboring point, the neighboring point exists in a rectangular shape. Thus, by associating neighboring points existing in the rectangular shape, for example, a “closed region” that is a region surrounded by ruled lines included in the prescription can be automatically recognized. These closed regions are, for example, information on the youngest number among four related numbers from among a series of numbers in the number column 511 among four consecutive ruled line information from a ruled line information column in the ruled line information data table 50. Define the information about the upper part of the closed area, the second young information as the information about the lower part of the closed area, the third young information as the information about the left edge of the closed area, and the fourth young information as the information about the right edge of the closed area. Thus, it becomes possible to associate the ruled line with the closed region. In addition, for example, even if the ruled line portion of the prescription that has been read is recognized as being broken due to, for example, the ruled line being missing or not being read correctly by a scanner, If it is less than a certain value, it may be determined that it is continuous and the closed region may be recognized. Alternatively, even when a rectangle is not formed by the combination of the four ruled lines, for example, it is assumed that one ruled line is interrupted and detected as two or more ruled lines, and the recognition of the flat area is performed. You can go. Heretofore, the area determination related to the ruled line information data table 50 has been described.

  As described above, after the information related to the ruled line of the read prescription is stored in the ruled line information data table 50, the prescription closed area information data table 60 is created from the ruled line information data table 50. The information data table 60 will be described. The prescription closed region information data table 60 includes a prescription closed region information management column 61 for managing the prescription closed region information data table, a closed region 1 information column 6201 for storing the read prescription closed region information, and a closed region 2 information column 6202. , Closed region 3 information column 6203, and so on. The ruled line closed region information management column 61 further includes a number column 611, a center X column 612, a center Y column 613, and the like, which are sequentially numbered 1, 2, 3, and the like. The prescription closed area information data table 60 is displayed on the display of the store output device 33 or information calculated from related information by the store device processor 32 according to a program stored in the store device memory 31, for example. According to the prescription closed area information data table setting screen (not shown), for example, information input by the operator from the keyboard of the store input device 34 of the store system 30 according to the program stored in the store apparatus memory 31, It is assumed that the information is stored on the store apparatus memory 31 by the store apparatus processor 31. The prescription closed area information data table setting screen (not shown) has the same format as the prescription closed area information data table 60. Further, if necessary, a prescription closed area information data table display screen (not shown) according to the contents of the prescription closed area information data table 60 is displayed on, for example, the store output device 33 of the store system 30 and the management system 20. It is displayed on the management output device 23. The prescription closed area information data table display screen (not shown) has the same format as the prescription closed area information data table 60. The prescription closed area information data table 60 may be created and / or input on the management system 20 side and stored in the memory. Further, the information input and stored on the sales office system 30 side may be transmitted to the management system 20 side via the network 10, for example, and created and / or input on the management system 20 side and stored in the memory. Information may be transmitted to the branch office system 30 side.

Regarding the prescription closed area information data table 60 configured as described above, in the prescription closed area information management column 61, for example, the number column 611 has “number”, the center X column 612 has “center X”, and the center Y, for example. For example, information “center Y” is stored in the column 613. Next, in the number column 611 such as the closed region 1 information column 6201, the closed region 2 information column 6202, the closed region 3 information column 6203,..., For example, “1”, “2”, “3”, Stores information such as. Next, the calculation method of each information stored in the center X column 612, the center Y column 613, etc. is shown below. For example, “13.743” in the center X column 612 of the closed region 1 information column 6201 corresponding to, for example, “1” stored in the number column 611 indicates an intermediate point between the ruled line end 411 and the ruled line end 412 in the X-axis direction. The average value of “4.229” stored in the start X column 512 of the ruled line 1 information column 5201 and “23.257” stored in the end X column 514,
(4.229 + 23.257) ÷ 2 = 13.743
The result of storing is shown.
Next, “20.086” in the center Y column 613, which is the median value in the Y-axis direction of the closed region, indicates an intermediate point between the ruled line end 411 and the ruled line end 413 in the Y-axis direction, and starts the ruled line 3 information column 5203. The average value of “15.857” stored in the X column 512 and “24.314” stored in the end X column 514,
(15.857 + 24.314) / 2 = 20.086
The result of storing is shown. Similarly, the median value of each closed region is also stored in the central X column 612 and the central Y column 613 of the closed region 2 information column 6202, the closed region 3 information column 6203,. Thus, the prescription closed area information data table 60 stores the median value of the read prescription closed area, and if the closed area information related to the prescription exists up to, for example, number 94, the prescription closed area information data table 60 stores the prescription closed area information data table 60. It shows that there are 94 closed regions. If no ruled line is detected or no closed region is recognized, for example, the entire prescription is recognized as one frame, and for example, the central X column 612 of the closed region 1 information column 6201 For example, information “0” is stored in the center Y column 613 to indicate that the closed region could not be recognized. The processing such as detection is performed by the store apparatus processor 32 according to a program developed on the store system 30, for example. The same processing may be performed by the management system 20. Further, the result may be transmitted to the sales office system 30 via the network 10, for example. In addition, the details of the prescription closed region information data table 60 are the same as the ruled line information data table 50, for example.

  On the other hand, in the present application, the format pattern of the prescription read by the prescription recognition system is stored in the data table for each new pattern including the sample format. Reading a new pattern may cause the prescription recognition system to read a pattern that is supposed to be read in advance, but the actual prescription that does not depend on the pattern that has been read in advance is also the data for each new pattern. A table is added and the information is stored in this data table. The prescription pattern data table 70 that stores information related to these patterns will be described with reference to FIG. The prescription pattern data table 70 stores prescription pattern fields 71 indicating which pattern information of the prescription, prescription pattern management field 72 for managing the prescription pattern data table, and information on the closed area included in the read prescription. A region 1 information column 7301, a closed region 2 information column 7302, a closed region 3 information column 7303, etc. are provided. In addition, a number field 721, a center X field 722, a center Y field 723, and an item content field 724 for storing information on what items each closed region has are configured. The prescription pattern data table 70 is, for example, information calculated from related information by the branch store processor 32 according to a program stored in the branch store memory 31 or a prescription pattern displayed on the display of the branch store output device 33. In accordance with a data table setting screen (not shown), for example, information input by an operator from the keyboard of the sales office input device 34 of the sales office system 30 according to a program stored in the sales office device memory 31, the sales office device processor It is assumed that the data is stored on the store apparatus memory 31 by 31. The prescription pattern data table setting screen (not shown) has the same format as the prescription pattern data table 70. Furthermore, if necessary, a prescription pattern data table display screen (not shown) according to the contents of the prescription pattern data table 70 is displayed on, for example, the store output device 33 of the store system 30 and the management output device 23 of the management system 20. Shall be displayed. The prescription pattern data table display screen (not shown) has the same format as the prescription pattern data table 70. The prescription pattern data table 70 may be created and / or input on the management system 20 side and stored in the memory. Further, the information input and stored on the sales office system 30 side may be transmitted to the management system 20 side via the network 10, for example, and created and / or input on the management system 20 side and stored in the memory. Information may be transmitted to the branch office system 30 side.

  Next, details of the information stored in the prescription pattern data table 70 will be described by taking information related to “prescription pattern A” as an example. First, in the prescription pattern column 71, for example, “prescription pattern: A”, for example, “number” in the number column 721 in the prescription pattern management column 72, and “item” in the item content column 724, for example, are shown. Each of the information “content” is stored. Next, in the number column 721 such as the closed region 1 information column 7301, the closed region 2 information column 7302, the closed region 3 information column 7303,..., For example, “1”, “2”, “3”, Are stored in the central X column 722, the central Y column 723, and the like. The calculation method of each information is the central X column 612 of the closed region 1 information column 6201 of the prescription closed region information data table 60, The median value of each closed region is stored in accordance with the calculation method in the center Y column 613. Thus, in the prescription pattern data table 70 as well as the prescription closed area information data table 60, the median value of the read prescription closed area is stored, and closed area information related to the prescription exists up to the number 94, for example. Indicates that there are 94 closed areas of the prescription. In this way, by recognizing the closed region and erasing the ruled line from the information (not shown) obtained by reading the prescription as described above, it is possible to assume a region for performing character recognition by the following OCR or the like. In addition to the effects described above, for example, when a closed region is erroneously recognized, and a portion recognized as a plurality of closed regions is actually one closed region, even if character recognition by OCR or the like is performed, the prescription In such a case, it is possible to read a certain character string by performing character recognition using OCR etc. including the periphery of the target closed area. It becomes. Further, in the item content field 724 of the closed area 1 information field 7301, for example, when a character previously entered in the closed area is read as a public expense bearer number when read by a character recognition method such as OCR, for example, In the item content column, for example, the information “Public bearer number” is stored. Similarly, information read by OCR, for example, is also stored in item content columns 724 such as a closed region 1 information column 7301, a closed region 2 information column 7302, a closed region 3 information column 7303,. Note that the character string stored in the item content column 724 is, for example, a list of items that are set in advance by the Ministry of Health, Labor and Welfare, the medical association, local public bodies, etc. It is also possible to improve reading accuracy by creating (not shown) and collating it. In this way, the prescription pattern data table 70 stores the median value of the closed area of the read prescription and the information related to the contents entered in the closed area, and, similarly to the prescription closed area information data table 60, the prescription. For example, if the closed area information regarding the number 94 is present, it indicates that there are 94 closed areas of the prescription. Further, as described above, when the closed region is not recognized and, for example, information “0” is stored in the center X column 612 and the center Y column 613 of the closed region 1 information column 6201, The read prescription raw image and the prescription pattern data table 70 are displayed in parallel on, for example, a display (not shown) connected to the store apparatus interface 36 of the store system 30, for example, an operator. Thus, the two items are compared and scrutinized. For example, if there is an error in the item contents stored in the item column 724, the operator corrects the item contents, and further, the item contents of the prescription pattern data table 70 are described. The position of the prescription is correlated with the position of the prescription from the raw image of the read prescription by pointing the screen with, for example, a touch pen. It may be subjected to a treatment to have prescription pattern data table. In addition, each process of the said detection, calculation, etc. including the said OCR is processed in the branch office apparatus processor 32 grade | etc., According to the program expand | deployed on the branch office system 30, for example, The result is carried out via the network 10, for example. It may be transmitted to the management system 20. The same processing may be performed by the management system 20. Further, the result may be transmitted to the sales office system 30 via the network 10, for example. In addition, regarding the details of the prescription pattern data table 70, for example, the prescription closed region information data table 60 and the ruled line information data table 50 are applied.

  As described above, the prescription pattern data table 70 for storing the information read for each prescription pattern has five prescription pattern data tables 70 when there are five types of prescription patterns, for example. In the prescription pattern column 71, information indicating which pattern is related, for example, “prescription pattern: A”, “prescription pattern: B”, “prescription pattern: C”,... Is stored. Then, by comparing the information about the format of the prescription identified in the past with the information of the prescription closed area information data table 60, it is confirmed whether or not the prescription read this time has already been registered as the format pattern. In this case, prescription according to the corresponding prescription pattern data table 70 Formatted as, when identifying the character information included in the closed area for example OCR or the like, may also be an item contents column 724 information such as the prescription pattern data table 70 as a reference. In addition, the format collation by the matching is performed in consideration of, for example, how many numbers in the number column 621 and the number column 721 exist, and how similar the numerical values related to the center of each closed region are. Furthermore, for example, by extracting features such as the aspect ratio, histogram, grid points, colors, etc. of the prescription closed region, and scoring them, and summing the points of the features that matched by the matching, which pattern It doesn't matter if it is close. Further, when the prescription read this time is based on a prescription pattern that has not been read so far, information on the prescription pattern is created as a new prescription pattern data table 70, for example, the store apparatus memory 31. Shall be stored above. However, regarding the registration of this new prescription pattern data table, for example, after displaying that it is a new prescription pattern data table (not shown) on the store output device 33 of the store system 30, the operator's It is good also as a process which stores after obtaining consent. It should be noted that the processing such as detection and calculation including the OCR and the details of storage in the data table are based on the ruled line information data table 50 and the prescription closed region information data table 60, for example.

  As described above, it is determined that the prescription read this time is the same as the prescription pattern read in advance, for example, and it is determined that the prescription pattern conforms to the prescription example diagram 40 shown in FIG. Next, the ruled lines are erased from the information (not shown) obtained by reading the prescription as described above, and the closed region surrounded by, for example, the ruled line end 411, the ruled line end 412, the ruled line end 4198, and the ruled line end 4199 is, for example, Assuming that the information follows the information stored in the closed area 1 information column 7301 of the prescription pattern data table 70 shown in FIG. 7, the public area bearer number is printed in the closed area. The prescription reading information data table 80 for storing information described in the prescription that matches the prescription pattern read in advance will be described with reference to FIG. The prescription reading information data table 80 includes a prescription reading information management column 81 for managing the prescription reading information data table, a prescription content 1 information column 8201 for storing each information of the read prescription, a prescription content 2 information column 8202, and a prescription content 3 An information column 8203, etc. are provided. Further, the prescription reading information management column 81 further includes a number column 811, numbered item number column 812, an item content column 812, a reading information column 813, and the like in order. The prescription reading information data table 80 is, for example, information calculated from related information by the branch store processor 32 in accordance with a program stored in the branch store memory 31 or a prescription displayed on the display of the branch store output device 33. According to the closed area information data table setting screen (not shown), for example, information input by the operator from the keyboard of the store input device 34 of the store system 30 is displayed in accordance with a program stored in the store apparatus memory 31. It is assumed that the store device processor 31 stores the information on the store device memory 31. The prescription reading information data table setting screen (not shown) has the same format as the prescription reading information data table 80. Further, if necessary, a prescription reading information data table display screen (not shown) according to the contents of the prescription reading information data table 80 is displayed on, for example, the store output device 33 of the store system 30 and the management output of the management system 20. It is displayed on the device 23. The prescription reading information data table display screen (not shown) has the same format as the prescription reading information data table 80. The prescription reading information data table 80 may be created and / or input on the management system 20 side and stored in the memory. Further, the information input and stored on the sales office system 30 side may be transmitted to the management system 20 side via the network 10, for example, and created and / or input on the management system 20 side and stored in the memory. Information may be transmitted to the branch office system 30 side.

  Regarding the prescription reading information data table 80 configured as described above, in the prescription reading information management column 81, for example, the number column 811 has “number”, the item content column 812 has “item content”, and the reading information column 813 has For example, information “read information” is stored. Next, in the number column 811 such as the prescription content 1 information column 8201, the prescription content 2 information column 8202, the prescription content 3 information column 8203,..., For example, “1”, “2”, “3”, .., And the item content column 812 stores the name of the item content stored in the read information column 813 described later. Specifically, for example, when the prescription read as described above conforms to the prescription pattern A, the information in the item content column 724 of the prescription pattern data table 70 regarding the prescription pattern A is stored. However, for example, the “patient” field shown in the patient information field 42 of FIG. 4 further includes specific items such as a “name” field and a “birth date” field. Although it is not necessary to store the information “patient”, for example, the item content column 812 of the prescription content 3 information column 8203 includes, for example, “patient name”, for example, the item content column 812 of the prescription content 4 information column 8204 includes, for example, Stores the information “patient birth date”. Similarly, the information is stored in the item content column 812 such as the prescription content 5 information column 8205, the prescription content 6 information column 8206,.

  As described above, when the item content corresponding to the item content column 812 of the prescription reading information data table 80 is stored, the ruled lines are erased from the information (not shown) read from the prescription in the reading information column 813 as described above. Analyzing information read by, for example, OCR of the written prescription, for example, the consistency between the item content and the nearby character string is used as a specific notation of the item content and the item content, for example A search is performed according to a data table (not shown) in which words or the like having a high possibility are associated with each other, character strings having affinity are extracted, and the result is stored. For example, in the prescription content 1 information column 8201, if the analyzed result is 6713128, for example, “6713128”, and in the prescription content 1 information column 8202, if the analyzed result is 1234567, for example, “1234567”, the prescription content 1 In the information column 8203, if the analyzed result is Taro Hongo, for example, information “Taro Hongo” is stored. Similarly, information is stored in each read information column 813 according to the analysis. Each processing such as ruled line detection and calculation from related information is processed by the branch office processor 32 according to a program developed on the branch office system 30, for example, and the result is transmitted via the network 10, for example. The data may be transmitted to the management system 20. The same processing may be performed by the management system 20. Further, the result may be transmitted to the sales office system 30 via the network 10, for example. In addition, regarding the details of the prescription reading information data table 80, for example, the ruled line information data table 50, the prescription closed area information data table 60, and the prescription pattern data table 70 are used.

  Next, a process for storing the information described in the closed area shown in the prescription column 43 of FIG. 4 in, for example, the reading information column 813 of the prescription content 21 information column 8221 of FIG. 8 will be described. “Prescription” is stored in the item content column 812 of the prescription content 21 information column 8221, but the most important prescription content in the prescription is read, analyzed, and stored in the read information column 813. An example of information described in the prescription column 43 of this prescription will be described according to prescription detailed information 90 shown in FIG. The prescription detail information 90 shows a part of the prescription column 43 shown in FIG. 4 and includes an unchangeable column 9431, a prescription detail column 9432, and the like. The prescription details column 9432 further includes specific prescription details, for example, 100 mg 94321, (100 mg 1 g) 94322, (dos not change dosage form) 94324,... Is done. For example, the underline and “drug”, “amount”, “supplement”, etc. under “Kefal Fine Granules for Children 100 mg” are added for the purpose of the following explanation, and are described in the actual prescription. Shall not. The reading of the prescription details column 9432 may be terminated by reading the margin 94329 below. As described above, the prescription details information 90 is, for example, a scanner (not shown) connected to the store input device 34 controlled by the store device processor 32 or the like according to a program developed on the store system 30, for example. The result is basically the same as the result of reading the prescription and the stored information (not shown) in the store device memory 31, for example.

  For character information relating to prescription details read as described above, first, for example, digital information read and analyzed by a scanner or the like is obtained, and then character information matching the digital information is converted into a digital information-character information correspondence database (not shown). OCR recognition processing is performed by performing in-region character recognition based on the corresponding character information extracted from (1). For example, with regard to drug names and usage, for example, via the network 10, for example, a database system related to the Ministry of Health, Labor and Welfare, which is a related institution system 202, a medical system related to various prefectures, a database system related to drug names by the Japanese Pharmacopoeia, etc. Connected with each system related to the prescription recognition system according to the above, for example, the branch office system 30 and stored in the database (not shown) of each system, etc., and the OCR recognition result Then, the reading accuracy as a prescription can be improved and the result can be stored in the digital information-character information correspondence database (not shown). In addition, as a process of selecting an attribute corresponding item, an attribute related to information estimated as item name information among character information obtained by character recognition in the region is estimated and / or specified, and the attribute and the attribute belong to the attribute An attribute-in-attribute possible item database corresponding to the estimated and / or specified attribute is selected from an attribute-in-attribute possible item database (not shown) associated with a list of possible item information. Further, as the similar item selection process, for the character information other than that related to the information estimated as the item name information among the character information obtained by the character recognition in the area, the attribute-attribute internal selected by the attribute corresponding item selection means It is also possible to select a similar item that selects items having a high degree of similarity among item information stored in a possible item database (not shown). The information read in this manner can be handled, for example, for a generic name drug, even for a drug name with a general display and a generic name code. As for the attribute-attribute-possible item database (not shown), as with the digital information-character information correspondence database (not shown), for example, a database system related to the managing director of the Ministry of Health, Labor and Welfare, which is a related institution system 202, Each system related to the prescription recognition system according to the present application, such as a medical system related to various prefectures, a database system related to drug names by the Japanese Pharmacopoeia, and a database of each system (not shown) It is also possible to collate with various kinds of information stored in the above and store the result in the attribute-attribute possible item database (not shown).

  Specifically, description will be made assuming that the prescription content desired to be entered in the prescription column 43 of the read prescription is as shown in the prescription details 90 shown in FIG. The prescription details 90 include a non-changeable column 9431, a prescription detail column 9432, and the like, and particularly the prescription column portion of the entire prescription shown in FIG. First, in the prescription detail column 9432, the indication “medicine” under the underline of 100 mg 94321 for kefral fine-grained children indicates that the 100 mg 94321 for kefal fine-grained children was recognized as a drug. ”,“ Titer ”,“ supplement ”,“ usage ”,“ days ”, and the like are also based on the result of collation with a database related to the Pharmaceutical Affairs Law, for example. Further, as will be described later, for example, (100 mg 1g) 94322 may be omitted so as not to be read.

  Next, in order to improve the processing speed for reading the prescription column 43 and / or improve accuracy, the prescription notation pattern list 100 that stores the specific prescription notation method defined for each prescription pattern in this way is described with reference to FIG. explain. First, the prescription notation pattern list 100 includes a prescription notation pattern management column 101 for managing the prescription notation pattern list, a prescription notation pattern 1 information column 10201, and a prescription notation pattern 2 information column 10202 for storing each information of the read prescription notifying method. , Prescription description pattern 3 information column 10203, and so on. The prescription notation pattern management column 101 further includes a prescription pattern column 1011, a reading format column 1012, and the like. The prescription notation pattern list 100 is, for example, information calculated from related information by the branch store processor 32 according to a program stored in the branch store memory 31 or a prescription notation displayed on the display of the branch store output device 33. According to a pattern list setting screen (not shown), for example, information input by an operator from the keyboard of the store input device 34 of the store system 30 according to a program stored in the store apparatus memory 31, the store apparatus processor It is assumed that the data is stored on the store apparatus memory 31 by 31. The prescription notation pattern list setting screen (not shown) has the same format as the prescription notation pattern list 100. Further, if necessary, a prescription notation pattern list display screen (not shown) according to the contents of the prescription notation pattern list 100 is displayed, for example, a store output device 33 of the store system 30 and a management output device 23 of the management system 20. Shall be displayed. The prescription notation pattern list display screen (not shown) has the same format as the prescription notation pattern list 100. The prescription notation pattern list 100 may be created and / or input on the management system 20 side and stored in the memory. Further, the information input and stored on the sales office system 30 side may be transmitted to the management system 20 side via the network 10, for example, and created and / or input on the management system 20 side and stored in the memory. Information may be transmitted to the branch office system 30 side.

  Regarding the prescription notation pattern list 100 configured as described above, for example, the prescription pattern column 1011 of the prescription notation pattern management column 101 stores information such as “prescription pattern” and the reading format column 1012 stores information such as “reading format”, respectively. . Next, the prescription pattern column 1011 such as the prescription notation pattern 1 information column 10201, the prescription notation pattern 2 information column 10202, the prescription notation pattern 3 information column 10203,... , “B”, “C”,..., And a specific prescription notation pattern is stored in the reading format column 1012. For example, as a prescription notation pattern of the prescription pattern A in the prescription notation pattern 1 information column 10201, for example, information “(drug / [potency] / amount / [supplement] / usage / days /) +” is stored. Similarly, the information is stored in the reading format column 1012 such as the prescription notation pattern 2 information column 10202, the prescription notation pattern 3 information column 10203,. The reading format stored in the reading format column 1012 is a format described in the corresponding prescription detail column 9432, which is read by, for example, OCR, and the format is identified and stored from the result. . For example, when there are a plurality of prescription pattern patterns for the prescription pattern A, that is, when the same prescription pattern A is used in, for example, the A hospital and the B hospital and the prescription pattern patterns are different from each other, for example, the prescription pattern Assume that “A” is stored in the prescription pattern column 1011 of the pattern 21 information column 10221, and for example, a prescription pattern different from the reading pattern of the prescription pattern 1 information column 10201 is stored in the reading format column 1012. Thereby, it is possible to cope with a case where a plurality of prescription writing patterns exist for the same prescription pattern.

  The information stored in this way indicates that it has the following prescription storage notation. For example, information “(drug / [potency] / amount / [supplement] / usage / number of days”) + is stored as a prescription pattern A of the prescription pattern A in the prescription pattern 1 information field 10201. Indicates that the item described first is the drug name, and the information “/” indicates the delimiter of the item, and so on. Next, “[titer]” indicates that the next item is the content of the titer, but “[]” surrounding the titer indicates that this item can be omitted. The omission can be determined, for example, by reading a prescription in which the item of titer is omitted while the same prescription pattern A is used, or by making an inquiry to the related institution system. . Similarly, the amount, [supplement], usage, and number of days are also indicated for each item and whether it can be omitted. Furthermore, the information "()" written at the beginning and end of the notation pattern item indicates that each item sandwiched between () is a series of prescriptions, and "+" is sandwiched between () It means that each item is repeated. Next, for example, in the prescription notation pattern of the prescription pattern B in the prescription notation pattern 2 information column 10202, the information “(drug // quantity /...)” Indicates that the first “drug” is the drug name as described above. The information “/” indicates an item delimiter, but “//” described below indicates that this item is always omitted in the corresponding pattern. The corner may be determined by, for example, the same prescription pattern B, but the corresponding item is always omitted, or by making an inquiry to the related institution system. The prescription pattern shall also indicate the notation pattern, and the prescription notation pattern should be a pattern that does not conform to the prescription notation pattern in the conventional prescription pattern. When the data is collected, it is added to the prescription notation pattern list 100. Each process is processed by the sales office device processor 32 or the like according to a program developed on the sales office system 30, for example. The result may be transmitted, for example, to the management system 20 via the network 10. Further, the same processing may be performed by the management system 20. Further, the result may be transmitted to the network 10, for example. May be transmitted to the sales office system 30. In addition, regarding details of the prescription notation pattern list 100, etc., for example, a ruled line information data table 50, a prescription closed area information data table 60, a prescription pattern data table 70, a prescription It shall be based on the reading information data table 80. Thus, the prescription notation pattern In the actual prescription reading process, the prescription reading pattern is first compared with the read prescription to match with a prescription pattern, and when a matching prescription pattern is found. Examines whether or not there are a plurality of prescription notation patterns in the same prescription pattern, for example, the contents of the prescription notation pattern list 100. For example, if one prescription notation pattern is registered, the details of the prescription according to its reading format In addition, when there are a plurality of prescription writing patterns, it is checked which of the plurality of prescription writing patterns matches. It is grasped by the total number of items in the reading format related to the prescription, or indicated by “[]” It may be determined by comparing the places where the optional items appear. Further, for example, the combination of the prescription pattern and the prescription notation pattern is managed by a list (not shown) for each medical institution information column 46 in FIG. 40 of the prescription example shown in FIG. Shall be read. In addition, when there is a problem such as a decrease in processing speed by referring to the prescription notation pattern list 100 due to a large number of corresponding prescription notation patterns, a process of referring to the prescription notation pattern list 100 Even when a new prescription notation pattern is read, new registration processing may not be performed at the operator's discretion, for example.

  As described above, according to the prescription example shown in FIG. 4 in the present application, the closed region is automatically recognized from the ruled line information, and the position of each prescription is determined by the positional relationship of each closed region and the OCR recognition of the character string in the closed region. The method of automatically extracting whether or not a hit is described, and then, as shown in FIG. 9, the method of acquiring an image obtained by cutting out the prescription column and performing OCR recognition has been described. In this application, it is possible to accurately recognize the OCR drug name, usage, etc., but the system can further make a more accurate determination by inquiring about the drug name, usage, etc. to the related institution system. To do. The result of reading the prescription is displayed in the same format as the finally read prescription. For example, the display is in the same format as the prescription illustrated in FIG. However, the finally read information is based on the result of recognizing the main information of the prescription as digital information including ruled lines, handwritten characters, and printed character information. The prescription is read and read by an image according to the result recognized as the digital information and, for example, a scanner (not shown) connected to the store input device 34 controlled by the store device processor 32 as described above. Information (not shown), for example, a raw image of a prescription stored in the store apparatus memory 31, for example, on a display (not shown) connected to the store apparatus interface 36 of the store system 30, etc. Displayed in parallel, for example, the operator compares both of them and carefully examines them. If there is an error in the digital information, the information shall be corrected and a separate data table (not shown) ) For example, in the store apparatus memory 31. Further, regarding the correction, it is possible to make correction possible even after the processing of recognizing the digital information is completed after reading the raw image of the prescription. Further, for example, when correction is performed by the operator, for example, prescription example diagram 40, ruled line information data table 50, prescription closed region information data table 60, prescription pattern data table 70, prescription reading information data table 80, processing of changing or adding the prescription notation pattern list 100 or the like may be performed. Furthermore, for example, in a certain sales office system, when the prescriptions to be handled are limited to medical institutions, etc., and the prescription pattern and prescription pattern are also limited, for example, the name of the medical institution and the prescription issued by the corresponding medical institution The correspondence between the prescription pattern and the prescription pattern is summarized in a data table (not shown). For example, the name of the medical institution described in the prescription is first read and recognized by OCR or the like, or the name of the medical institution is determined by the operator. For example, by inputting from a keyboard or the like of the store input device 34 of the store system 30, first, it may be read as a prescription with a specific prescription pattern and prescription pattern. Each processing relating to the parallel display and correction described above is processed by the branch store processor 32 or the like, for example, according to a program developed on the branch store system 30, and the result is managed by the management system 20 via the network 10, for example. You can send it to The same processing may be performed by the management system 20. Further, the result may be transmitted to the sales office system 30 via the network 10, for example.

  Next, the prescription recognition system process outline diagram 110 in the present application will be described with reference to FIG. The prescription recognition system process overview diagram 110 includes an FGS 111, form information 112, FES 113, HWS 114, ARS 115, and HCS 116. Hereinafter, the configuration of the FGS 111 and the FES 113 and the processing content of the FGS 111 will be described with reference to the prescription recognition system process overview diagram 110.

First, the processing content of the FGS 113 will be described. The FGS includes the following three processes. First, regarding reading a new prescription, a new screen file input waiting 1111 and form information reading / update 1112 are performed, and then matching 1113 with form information is performed.
Less than,
0. Inclination correction processing is performed by the inclination correction 11131. Extraction and comparison of image features by image feature extraction 11132: Image features are extracted from the image and compared with features of a known form.
2. Form information extraction 11133 recognizes ruled lines, typefaces, and the like from the image, and creates data that is the basis of the editing data of the FES 113.

Next, each processing flow and considerations are shown.
0. Inclination correction This process is performed as a pre-process. Since it is performed for all images, the processing is reduced as much as possible. Although the example of the horizontal line part extracted for inclination correction by the process of this application is not an application example with respect to a prescription, it is shown in FIG.
<Horizontal line extraction>
The processing contents are as follows.
Assuming that line segments such as frame lines are much darker than the image background, search for dark pixels in the horizontal direction.
-A part where a certain number or more of dark pixels continue in the horizontal direction is assumed to be a horizontal line segment, and its start end coordinates are acquired.

  Next, as shown in FIG. 13, the sum of the pixel values on the line segments 13021 to 13023 obtained by changing the angle of the line segment 1302 (not shown) is repeatedly obtained from the start end coordinates 1303 of the frame line 1301, and the frame line The angle at which the sum of the pixel values that do not overlap 1301 and the line segments 13021 to 13023 becomes the smallest (darker) is recorded.

  Such processing is performed for a plurality of horizontal line segments, and the calculated median value of the angle group is set as a correct angle. There is an advantage that the processing becomes relatively light. However, when the width of the line segments 14021 to 14022 is narrower than the line width of the frame line 1401, the correct angle may not be obtained. This pattern is shown in FIG. As a countermeasure, the line width of the frame line is detected, and the shape of the line segment 1402 (not shown) is changed.

As an alternative to the above, there is the following technical idea.
<Using dispersion (There is no application example for a prescription, but the outline of the technical idea is shown in FIGS. 15 and 16)>
As processing contents, the total of the pixel values in the horizontal direction is calculated for each row. Since the ruled lines are dark, there are few undulations when there is an inclination.
in this case,
The square distance is calculated for the adjacent horizontal pixel total values on the assumption that the undulation of the total pixel values of each row becomes large in a horizontal image using the total of the pixel values of each row as an element.
-Rotate the image slightly and repeat the above process to make the angle at which the square distance is maximum the correct angle.
Do that. By doing this, the same tendency appears for horizontally written character strings, and therefore, there is an advantage that it can be applied to a document image without ruled lines. There is also a problem that processing is heavy. When correcting a large inclination, the processing time becomes long. Therefore, an image that is extremely inclined is not dealt with, and the operator corrects it with the FES 113 and sends it to the FGS 111 again.
As long as the experiment was conducted by the applicant, the inclination falls within 1 ° when scanned correctly.

<Extraction and comparison of image features>
In order to reduce the comparison processing, a plurality of image features are acquired.
1-1. Candidate narrowing image feature When matching processing is performed with all known forms, the processing time is expected to be very long. Therefore, this is used to narrow down the known forms of the target candidates.
-Image features for alignment, values that can be extracted from matching features (average value, variance, etc.)
・ Aspect ratio of document ・ Ratio of horizontal and vertical components ・ Number of grid points If it takes time to extract features for refinement, there is no original or child. .

1-2. Image features for alignment (although there is no application example for a prescription, FIG. 17 shows an outline of the technical idea)
As a pre-matching process, it is used to match the position of the feature value calculation for matching between the image used for feature extraction of the known form and the input image.
-Total pixel value in the horizontal direction for each row-Total pixel value in the vertical direction for each column

1-3. Image features for matching Used to compare a known form with the current form. For each known form to be compared, alignment is performed and re-generation is performed. In the FES113 process, the following procedure is used in order to obtain a feature that can leave a comparison result that can visually display where the dissimilar part is located on a known form that is matched. Do.
First, the image features for matching known forms are automatically divided into a fixed number of X-axis 1801 and Y-axis 1802 when the form is registered in the FES 113, and each divided area except for the handwritten areas 1803-1805 is used. Extract and register image features. (See Figure 18)

  Next, with respect to the input image, the division region is adjusted so as to match the matching feature of the known form to be compared, using the image feature for alignment. If the divided area is deviated by more than a certain value (1901) or distorted, it is determined as NG and the comparison target is moved to the next known form. Finally, the image characteristics for matching in each divided area are acquired and compared. (See Figure 19)

As a result, the matching result is displayed, for example, on the screen as follows.
When compared with the correct form: A small number of each of the divided areas having different matching image features is displayed.
When compared with different forms: A large number of the divided areas having different matching image features are displayed.
Note that the display of each divided region is made recognizable at a glance, for example, by a color that is easy to recognize.

As mentioned above, when form recognition fails or when a similar form is used, the part to be visually confirmed is intuitively understood. The image features for matching are selected from the following two processes and processed.
・ Average value of each RGB color ・ Pattern of edge (luminance difference between adjacent pixels) In other words, in the case where only the information in the handwritten area needs to be acquired by the system, type / ruled line information is not required. No input is made. In that case, there is no data to be superimposed and displayed when the confirmation screen of the FES 113 is displayed. To avoid this, the image used for registration of the known form is left for confirmation / superimposition display.

  As described above, the prescription has been described as an example. However, since the present application makes it possible to accurately recognize the characters written on the prescription, not only medical treatment and dentistry, but also other than prescriptions, such as trade transactions It has applicability in all industries that use characters entered in standard forms such as official documents.

Claims (15)

An area determination unit for recognizing ruled lines and / or frame information of a document including a form, and determining an area related to the recognized ruled lines and / or frame information;
For a specific area among the areas determined by the area determination means, characters in the area are imaged and reduced to digital information, and character information that matches the reduced digital information is digital information-character information correspondence An in-region character recognition unit that extracts corresponding character information from the database;
Estimates and / or defines attributes related to item name information and estimated information among character information obtained by the in-region character recognition means, and associates attributes with a list of item information that can be assumed to belong to the attributes An attribute corresponding to the estimated and / or specified attribute from the attribute-possible attribute item database, and an attribute corresponding item selection unit for selecting the attribute-possible attribute database;
Of the character information obtained by the in-region character recognition means, character information other than that associated with the information estimated as item name information is stored in the attribute-attribute possible item database selected by the attribute corresponding item selection means. A similar item selection section for selecting items with a high degree of similarity,
A recognition apparatus comprising: a character information determination unit that verifies validity of the item information with high similarity selected by the similar item selection unit and determines information corresponding to the character information. A frame line recognition device for recognizing ruled lines and / or frame information of a fixed form document including a form;
An area determination means for determining an area related to ruled lines and / or frame information recognized by the frame line recognition device;
For a specific area among the areas determined by the area determination means, characters in the area are imaged and reduced to digital information, and character information that matches the reduced digital information is digital information-character information correspondence In-region character recognition means for extracting corresponding character information from the database;
Estimates and / or defines attributes related to item name information and estimated information among character information obtained by the in-region character recognition means, and associates attributes with a list of item information that can be assumed to belong to the attributes An attribute corresponding item selection means for selecting an attribute corresponding to the estimated and / or specified attribute from the attribute-possible attribute item database and the attribute-possible attribute item database;
Of the character information obtained by the in-region character recognition means, character information other than that associated with the information estimated as item name information is stored in the attribute-attribute possible item database selected by the attribute corresponding item selection means. Similar item selection means for selecting items with high similarity among the item information,
A form recognition system comprising: character information confirmation means for verifying validity of the item information having a high similarity selected by the proximity item selection means and confirming information corresponding to the character information. An area determination step for recognizing ruled lines and / or frame information of a document including a form and determining the area related to the recognized ruled lines and / or frame information;
For a specific area of the determined area, the characters in the area are imaged and reduced to digital information, and character information that matches the reduced digital information is extracted from the digital information-character information correspondence database. The character recognition step in the area to obtain the corresponding character information,
Attribute-attribute that estimates and / or defines an attribute related to information estimated as item name information among the obtained character information, and associates the attribute with a list of item information that can be assumed to belong to the attribute An attribute corresponding to the estimated and / or prescribed attribute from the internal possible item database-an attribute corresponding item selecting step for selecting an internal attribute possible item database;
Of the item information stored in the attribute-attribute-possible item database selected by the attribute-corresponding item selection means for character information other than that relating to information estimated as item name information among the obtained character information Proximity item selection step for selecting items with high similarity,
A form recognition method comprising: verifying validity of the selected item information with high proximity and determining information corresponding to the character information. Computer
Area determining means for recognizing ruled lines and / or frame information of a fixed form document including a form and determining an area related to the recognized ruled lines and / or frame information;
For a specific area among the areas determined by the area determination means, characters in the area are imaged and reduced to digital information, and character information that matches the reduced digital information is digital information-character information correspondence In-region character recognition means for extracting corresponding character information from the database;
Estimates and / or defines attributes related to item name information and estimated information among character information obtained by the in-region character recognition means, and associates attributes with a list of item information that can be assumed to belong to the attributes An attribute corresponding item selection means for selecting an attribute corresponding to the estimated and / or specified attribute from the attribute-possible attribute item database and the attribute-possible attribute item database;
Of the character information obtained by the in-region character recognition means, character information other than that associated with the information estimated as item name information is stored in the attribute-attribute possible item database selected by the attribute corresponding item selection means. Proximity item selection means for selecting the item information with high similarity,
The form comprising: operating as character information confirmation means for verifying validity of the item information with high similarity selected by the proximity item selection means and confirming information corresponding to the character information Recognition program.   A recording medium on which the program according to claim 4 is mounted. Regarding the format of the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit,
Means for changing the recognized form pattern and prescription notation pattern;
Means for associating a plurality of prescription patterns with the same form pattern;
Means for storing information relating to the associated results as a first storage;
The form recognition apparatus according to claim 1, further comprising: means for displaying information stored as the first storage. Regarding the format of the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit,
Means for changing the recognized form pattern and prescription notation pattern;
Means for associating a plurality of prescription patterns with the same form pattern;
Means for storing information relating to the associated results as a first storage;
The form recognition system according to claim 2, further comprising: means for displaying information stored as the first storage. Regarding the format of the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit,
A step for changing the recognized form pattern and prescription notation pattern;
Associating a plurality of prescription patterns with the same form pattern;
Storing information on the associated result as a first storage;
The form recognition method according to claim 3, further comprising: displaying information stored as the first memory. Computer
Regarding the format of the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit, and the character information determination unit,
Means for changing the recognized form pattern and prescription notation pattern;
Means for associating a plurality of prescription patterns with the same form pattern;
Means for storing information relating to the associated results as a first storage;
The form recognition program according to claim 4, further comprising operating as means for displaying information stored as the first storage.   A recording medium carrying the program according to claim 9. Regarding the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit and the character information determination unit,
Means for displaying at least one of the results of recognizing ruled lines, handwritten characters and printed character information as digital information;
Means for modifying the displayed digital information;
Means for storing information relating to the corrected result as a second storage;
The form recognition apparatus according to claim 6, further comprising: means for displaying information stored as the second storage. Regarding the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit and the character information determination unit,
Means for displaying at least one of the results of recognizing ruled lines, handwritten characters and printed character information as digital information;
Means for correcting the displayed digital information;
Means for storing information relating to the corrected result as a second storage;
The form recognition system according to claim 7, further comprising: means for displaying information stored as the second storage. Regarding the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit and the character information determination unit,
Displaying at least one of the results of recognizing ruled lines, handwritten characters and printed character information as digital information;
Modifying the displayed digital information;
Storing information on the corrected result as a second memory;
The form recognition method according to claim 8, further comprising: displaying information stored as the second storage. Computer
Regarding the form recognized by the region determination unit, the in-region character recognition unit, the attribute corresponding item selection unit, the similar item selection unit and the character information determination unit,
Means for displaying at least one of the results of recognizing ruled lines, handwritten characters and printed character information as digital information;
Means for modifying the displayed digital information;
Means for storing information relating to the corrected result as a second storage;
The form recognition program according to claim 9, further comprising operating as means for displaying information stored as the second storage.   A recording medium on which the program according to claim 14 is mounted.

Images