JP6165595B2 - Form identification system - Google Patents

Description

  Embodiments described herein relate generally to a form identification system.

  A technique for registering data as electronic data by classifying the character recognition result of an image acquired from a form into items such as name, address, and telephone number is known. In this type of data registration, the type of form to be registered, that is, the form of the form that can obtain the correspondence between each character string on the form and each item name is generally specified before character recognition. There is a need.

  For example, in a system composed of a client terminal having a function of acquiring a form image and performing character recognition and a management server having a function of specifying a form type (style), first of all, In order to specify the type, an image of the entire form is transmitted from the client terminal side to the management server side.

JP-A-2005-216102

  However, since the image of the entire form has a large data capacity, the communication path band between the client terminal and the management server is compressed, and communication takes a long time. Also, the image data processing on the management server side for specifying the form type has a large processing load, and the load also increases in proportion to the number of client terminals. For this reason, image data processing on the management server side becomes a bottleneck in the entire system.

  Accordingly, the problem to be solved by the present invention is to provide a form identification system that can reduce the amount of communication on the network and reduce the data processing load on the management server side.

The form identification system according to the embodiment includes a client terminal and a management server connected via a network. The client terminal includes an image acquisition unit, a feature point information extraction unit, and a transmission unit. The image acquisition unit acquires a form image. The feature point information extraction unit extracts feature points of the form as feature point information from the acquired image. The transmission unit transmits the extracted feature point information. Management server includes receiving unit, the identification processing unit and a second transmission unit. The receiving unit receives the transmitted feature point information. The identification processing unit performs identification processing of the form type based on the received feature point information. The second transmission unit transmits request information for requesting designation of a predetermined element from among a plurality of elements defining the form type. Furthermore, the client terminal includes a second reception unit, a screen generation unit, and an element selection unit. The second receiving unit receives the transmitted request information. The screen generation unit generates a guidance screen that guides the designation of the predetermined element based on the received request information. The element selection unit selects an element that has been designated and operated according to the guidance from the plurality of elements. Further, the feature point information extraction unit included in the client terminal extracts feature point information corresponding to the selected element from the acquired form image.

The figure which shows typically the structure of the form identification system which concerns on 1st Embodiment. The figure which shows typically the structure of the form identification system of a comparative example. The figure which shows typically the main processes by the form identification system of FIG. The block diagram which shows the main hardware constitutions of the form identification system of FIG. The block diagram which shows the structure of the form identification system of FIG. 1 functionally. The figure which illustrated the structure of form image DB with which the client terminal of the form identification system of FIG. 5 was equipped. The figure which illustrated the composition of form definition DB with which the management server of the form identification system of Drawing 5 was provided. The figure which illustrated the contents of feature point TBL which constitutes a part of form definition DB of Drawing 7. The figure which shows the structure of the feature point item TBL linked | related with the feature point TBL of FIG. The figure which shows the structure of the form definition body TBL linked | related with the feature point TBL of FIG. 6 is a flowchart showing general processing by the form identification system in FIG. 5. FIG. 6 is a sequence diagram showing a form photographing process by the form identification system in FIG. 5. FIG. 6 is a sequence diagram showing form candidate selection processing by the form identification system of FIG. 5. The sequence diagram which shows the form definition body specific process by the form identification system of FIG. The functional block diagram which shows the structure of the form identification system which concerns on 2nd Embodiment. The figure which shows typically the process by the form identification system of FIG. The block diagram which shows functionally the structure of the form identification system which concerns on 3rd Embodiment. The figure which shows typically the process by the form identification system of FIG. The figure which illustrated the structure of feature point item DB with which the client terminal of the form identification system of FIG. 17 was provided. The block diagram which shows functionally the structure of the form identification system which concerns on 4th Embodiment. The figure which shows typically the process by the form identification system of FIG. The block diagram which shows roughly the structure of the form identification system which concerns on 5th Embodiment. The figure for demonstrating the structure of the form identification system which concerns on 6th Embodiment, and the usage example of a form definition body.

Hereinafter, embodiments will be described with reference to the drawings.
<First Embodiment>
As shown in FIG. 1, the form identification system 40 according to the first embodiment includes a client terminal 50 and a management server 60 connected via a network.

  As shown in FIG. 1, the form identification system 40 according to the present embodiment transmits feature point information extracted from the form image from the client terminal 50 side to the management server 60 side. In addition, the form identification system 40 transmits a form definition body (information defining a format for each form type) specified mainly based on feature point information from the management server 60 side to the client terminal 50 side.

  On the other hand, as shown in FIG. 2, the form identification system of the comparative example transmits an image of the entire form from the client terminal side to the management server side, and the form definition body specified by processing the image of the entire form is managed by the management server. From the side to the client terminal side.

  In other words, the form identification system 40 of the present embodiment reduces the data processing load on the management server 60 side by performing part of the data processing on the form image on the client terminal 50 side. In addition, the form identification system 40 converts the information transmitted from the client terminal 50 side to the management server 60 side into feature point information having a relatively small data capacity, not an image of the entire form having a large data capacity as in the comparative example. As a result, the amount of data communication between the client terminal 50 and the management server 60 is reduced.

  Further, as shown in FIG. 3, the form identification system 40 uses the GUI (Graphical User Interface) on the client terminal 50 side to obtain feature point information sequentially obtained from the image of the form 1 from the client terminal 50 side. By transmitting to the 60 side one by one, the type (form) of the form 1 is specified or narrowed down on the management server 60 side.

  Here, the main hardware configuration of the form identification system 40 will be described. As shown in FIG. 4, the client terminal 50 and the management server 60 are connected to each other via a network 2 such as a WAN (Wide Area Network).

  The client terminal 50 is a portable terminal such as a smart phone or a tablet provided with a touch screen display or a personal computer provided with a pointing device such as a mouse. As shown in FIG. 4, the client terminal 50 includes a camera unit 4, an LCD display unit 5, a key operation unit 3, a CPU 7, a storage unit 8, and a communication unit 9. The camera unit 4 captures the form 1 and obtains an image of the form 1. The LCD display unit 5 displays a screen. The key operation unit 3 receives an input operation from a button or a keyboard displayed on the screen. The CPU 7 performs software processing. The storage unit 8 stores an image of the form 1 and software. The communication unit 9 communicates with the management server 60 side via the network 2.

  On the other hand, the management server 60 is configured by a computer that collectively manages all client terminals existing on the network 2. As shown in FIG. 4, a communication unit 10 that communicates with the client terminal 50 side via the network 2, a CPU 11 that performs software processing, and a storage unit 12 that stores a form definition body and software are provided.

  Next, functional configurations of the client terminal 50 and the management server 60 will be described. As shown in FIG. 5, the client terminal 50 includes an image acquisition unit 13, a display function unit 14, an operation function unit 15, a feature point extraction function unit 16, a communication function unit 17, and a form image DB (form image database) 21. I have.

  As shown in FIG. 5, the image acquisition unit 13 includes a camera unit 4 and a form image writing unit 23. That is, the image acquisition unit 13 acquires an image of the form 1 captured through the camera unit 4. The form image writing unit 23 stores the captured image of the form 1 in the form image DB 21 as a form image. The display function unit 14 includes an LCD display unit 5, a screen generation unit 25, and a form image reading unit 24. The form image reading unit 24 reads the form image stored in the form image DB 21. The screen generation unit 25 generates a screen for displaying image data such as a form image and a form candidate list (form type selection candidate list). The LCD display unit 5 displays the screen generated by the screen generation unit 25.

  As shown in FIG. 5, the operation function unit 15 includes a key operation unit 3, a feature point coordinate designating unit 26 having a function as an element selection unit, and a form designating unit 27. The key operation unit 3 described above receives an input operation by the user. The feature point coordinate designating unit 26 selects an element designated and operated by the user via the key operation unit 3 from a plurality of elements defining the type (style) of the form. Specifically, as shown in FIG. 3, the feature point coordinate designating unit 26 designates a designated and operated area (for example, a predetermined activated via a GUI) in the image of the form 1 displayed on the LCD display unit 5. Select (specify) the coordinate position of the item name description area). The form designating unit 27 selects (specifies) the form type that has been designated and operated from, for example, the above list of form candidates (form type selection candidate list).

  As shown in FIG. 5, the feature point extraction function unit 16 includes a feature point information extraction unit 28. The feature point information extraction unit 28 extracts the feature points of the form 1 from the image of the form 1 acquired by the image acquisition unit 13 as feature point information. Specifically, the feature point information extraction unit 28 is assigned to the region (selected coordinate position) designated and operated in the image of the form 1 by the feature point coordinate designating unit 26 and the region on the form. Information relating item names and the like is extracted as feature point information.

  The communication function unit 17 transmits and receives data between the client terminal 50 and the management server 60. The communication function unit 17 includes a feature point information transmission unit 29, a form candidate list reception unit 30, a form ID transmission unit 31, and a form definition body reception unit 32 that function as a transmission unit (first transmission unit).

  The feature point information transmission unit 29 transmits the extracted feature point information to the management server 60. The form candidate list receiving unit 30 receives a form candidate list (form type selection candidate list) from the management server 60. The form ID transmission unit 31 transmits a form ID (type identification information unique to each form type) corresponding to the form type designated by the form designation unit 27 to the management server 60. The form definition body receiving unit 32 receives the form definition body from the management server 60.

  Here, the form definition body is form definition information in which a form format (format) unique to each form type is defined. The form definition body is also referred to as a form model or form form. In other words, the form definition body includes each item name existing on the form, the coordinate position (area) on the form to which the description area for each item name is assigned, a unique form ID and form type for each form type. This is form definition information (data file) in which size and the like are associated (linked) with each other. Note that the form definition received by the form definition receiving unit 32 is another function of the client terminal 50 (for example, various types of forms as shown in the diagram for explaining the use example of the form definition in FIG. 23). A form character recognition function that recognizes the image part in the entry area of the item name that requires data extraction from the form image and registers the character data [each text] that is the character recognition result as an item. Etc.).

  On the other hand, as shown in FIG. 5, the management server 60 includes a communication function unit 18, a form identification function unit 19, a form definition body reading function unit 20, and a form definition body DB (form definition body database) 22. The communication function unit 18 transmits and receives data between the client terminal 50 and the management server 60. The communication function unit 18 includes a feature point information reception unit 33, a form candidate list transmission unit 34, a form ID reception unit 35, and a form definition body transmission unit 36 that function as a reception unit (first reception unit).

  The feature point information receiving unit 33 receives feature point information from the client terminal 50. The form candidate list transmission unit 34 transmits the form candidate list to the client terminal 50. The form ID receiving unit 35 receives the form ID from the client terminal 50. The form definition body transmission unit 36 transmits the form definition body to the client terminal 50.

  The form identification function unit 19 includes an identification processing unit 37. The identification processing unit 37 performs identification processing of the form type based on the received feature point information. Specifically, the identification processing unit 37 performs identification processing while referring to the form definition body DB 22. This identification process is a process for narrowing down the types of forms into several types (a process for creating a form candidate list), or a process for specifying the type of form as one type. The form definition body reading function unit 20 includes a form definition body reading unit 38. The form definition body reading unit 38 reads the form definition body corresponding to the form ID received by the form ID reception unit 35 from the form definition body DB 22.

  Here, the client terminal 50 and the management server 60 are equipped with various types of hardware such as a main memory such as a RAM, an auxiliary (secondary) storage device such as an HDD, a CPU, and a ROM. The client terminal 50 and the management server 60 are functionally shown in FIG. 5 by loading a client terminal program and a management server program stored in advance in an auxiliary storage device, a ROM, and the like on each main memory. The main components are realized by software, for example. Note that these components may be configured by hardware instead of software.

  The form image DB 21 described above stores a photographed image ID and form image data in association with each other as shown in FIG. Further, as shown in FIG. 7, the form definition DB 22 includes a form ID, feature point information (for example, item name [item name such as title, name, address, etc.) and coordinate position [x1, y1 etc. Information associated with [position]) and the form definition body are stored in association with each other. To illustrate a more specific configuration, the form definition DB 22 includes, for example, feature points TBL (feature point table) 22a associated (linked) with each other with a predetermined ID, as shown in FIGS. A feature point item TBL (feature point item table) 22b and a form definition body TBL (form definition body table) 22c are included.

  As shown in FIG. 8, the feature point TBL 22a includes a form ID, a feature point item ID, and position information (X coordinate, Y coordinate, height, and width) of a predetermined description frame (description area) on the form. (Position information) and possible values (values indicating the types of corresponding forms) are stored in association with each other. As shown in FIG. 9, the feature point item TBL 22b stores a feature point item ID and item names such as title, name, address, and age in association with each other. Further, as shown in FIG. 10, the form definition TBL22c stores a form ID and the electronic file name of the form definition in association with each other. That is, the feature point item TBL22b is linked to the feature point TBL22a by the feature point item ID, while the form definition TBL22c is linked to the feature point TBL22a by the form ID. As a result, one or more pieces of positional information (feature point information) including X coordinates, Y coordinates, height, width, etc. in a predetermined description frame on the form extracted from the form image are obtained. It is possible to specify a form definition body corresponding to the above or narrow down candidates for the form definition body.

  FIG. 11 is a flowchart showing general processing by the form identification system 40. In the form identification system 40, as shown in FIG. 11, after the form is photographed (S1), feature point information on the form image is extracted according to a predetermined input operation (S2). If there is a form candidate list (form type selection candidate list) corresponding to the extracted feature point information that corresponds to the photographed form (YES in S3), a selection operation (designation operation) from the form candidate list ), A form definition body corresponding to the type of photographed form is transmitted from the management server 60 to the client terminal 50 (S4).

  The above-described form candidate list may be a list of character strings such as an automobile inspection certificate, an insurance policy, and a license. The form definition received on the client terminal 50 side is used later when executing another function of the client terminal 50 (such as a form character recognition function for registering character recognition results as data) (S5).

  Next, the form photographing process, form candidate selection process, and form definition body specifying process by the form identification system 40 will be described based on the sequence diagrams shown in FIGS. As shown in FIG. 12, in the form photographing process, first, the camera unit 4 photographs the form 1. The form image writing unit 23 writes the captured image of the form 1 into the form image DB 21 as a form image. In response to a display request for a form image by an input operation from a user (user), the form image writing unit 23 reads a corresponding form image from the form image DB 21. The screen generation unit 25 generates a display screen for displaying the read form image, and displays the display screen on the LCD display unit 5.

  As shown in FIG. 13, in the form candidate selection process, the feature point coordinate designating unit 26 designates the coordinate position designated (selected) by the user (user) on the form image from the key operation unit 3 on the GUI. The acquired coordinate position is passed to the feature point information extraction unit 28. The feature point information extraction unit 28 controls the form image reading unit 24 to read the form image, the acquired coordinate position on the form image, the item name on the form assigned to the coordinate position, and the like. And feature point information associated with.

  The feature point information transmitting unit 29 transmits the feature point information transferred from the feature point information extracting unit 28 to the feature point information receiving unit 33 on the management server 60 side. The identification processing unit 37 performs identification processing of the form type based on the feature point information transferred from the feature point information receiving unit 33. Specifically, the identification processing unit 37 refers to the form definition DB 22 and obtains a form candidate list (form type candidate list) from the form definition corresponding to the received feature point information.

  As illustrated in FIG. 13, the form candidate list transmitting unit 34 transmits the form candidate list delivered from the identification processing unit 37 to the form candidate list receiving unit 30 on the client terminal 50 side. The screen generation unit 25 generates a display screen for displaying the form candidate list received by the form candidate list receiving unit 30 as image data. The LCD display unit 5 displays a generated display screen including a form candidate list.

  Furthermore, as shown in FIG. 14, in the form definition body specifying process, the key operation unit 3 performs an operation (input) of a desired candidate performed by the user (user) via the GUI from the form candidate list on the display screen. Operation). The form designating unit 27 obtains a form ID corresponding to the candidate that has been designated. The form ID transmitting unit 31 transmits the form ID notified from the form specifying unit 27 to the form ID receiving unit 35 on the management server 60 side.

  As shown in FIG. 14, the form definition body reading unit 38 acquires the form definition body corresponding to the form ID from the form definition body DB 22 using the form ID received by the form ID reception unit 35 as a search key. The form definition body transmission unit 36 transmits the form definition body DB 22 acquired by the form definition body reading unit 38 to the form definition body reception unit 32 on the client terminal 50 side.

  As described above, according to the form identification system 40 according to the first embodiment, the information transmitted from the client terminal 50 side to the management server 60 side is not an image of the entire form having a large data capacity, but the data capacity. By using relatively small feature point information, the communication capacity between the client terminal 50 and the management server 60 can be reduced. Further, according to the form identification system 40, a part of the data processing for the form image is performed on the client terminal 50 side (data processing for extracting feature points from the form image) is performed on the management server 60 side. Can reduce the data processing load.

<Second Embodiment>
Next, a second embodiment will be described based on FIGS. 15 and 16. In FIG. 15, the same components as those in FIG. 5 described in the first embodiment are given the same reference numerals, and redundant descriptions are omitted. As illustrated in FIG. 15, the form identification system 41 according to the present embodiment replaces the client terminal 50 and the management server 60 included in the form identification system 40 according to the first embodiment with a client terminal 51 and a management server 61. Have

  As illustrated in FIG. 15, the client terminal 51 replaces the form candidate list reception unit 30 and the screen generation unit 25 provided in the client terminal 50 with a request information reception unit 30 a as a second reception unit, and a screen. A generation unit 25a is provided. The management server 61 includes a request information transmission unit 34a and an identification processing unit 37a as a second transmission unit in place of the form candidate list transmission unit 34 and the identification processing unit 37 provided in the management server 60.

  The request information transmission unit 34a on the management server 61 side transmits request information for requesting designation of a predetermined element from among a plurality of elements that define the type (form) of the form. On the other hand, the request information receiving unit 30a on the client terminal 51 side receives the request information transmitted from the management server 61 side. The screen generation unit 25a generates a guidance screen that guides the designation of a predetermined element based on the received request information. The feature point coordinate designating unit 26 having a function as an element selecting unit allows the user to select the key operation unit 3 according to the guidance (guide) on the guidance screen from the plurality of elements defining the form type. Select the specified element via The feature point information extraction unit 28 extracts feature point information corresponding to the selected element from the image of the form 1 acquired by the image acquisition unit 13.

  Specifically, the form identification system 41 according to the present embodiment can interactively specify elements for obtaining feature point information through the GUI, and can more efficiently narrow down the number of candidates in the form candidate list. In order to do so, the management server 61 requests the client terminal 51 to specify an element (designation operation of a coordinate position of a description area in a predetermined item name). In the first embodiment, the identification processing unit 37 simply outputs the form candidate list. However, in order to narrow down the number of candidates in the form candidate list most efficiently, the identification processing unit 37a should be designated. An element (item name) is selected, and this is notified to the request information transmission unit 34a. The screen generation unit 25a generates a guidance screen including an instruction message about an element (item name) for which a designation operation is requested.

  Next, a process of narrowing down the form type (style / type) by exchanging information between the client terminal 51 and the management server 61 will be described with reference to the flowchart shown in FIG.

  Based on the instruction from the management server 61, the LCD display unit 5 of the client terminal 51 displays a guidance screen including an instruction message “tap the title” (S11). In response to this guidance, an area of coordinates (x1, y1) where the title on the image of the form 1 is located is designated (S12).

  The management server 61 searches the form definition body DB for the form definition body (form ID) whose title is at the coordinates (x1, y1) (S13). The management server 61 (identification processing unit 37a) uses the form A (form A) and the form B (form B) as candidates, and further analyzes the difference between the form A and form B to obtain the most efficient. The client terminal 51 side is requested for the item name to be specified in order to narrow down the types of forms well (S14).

  Since the coordinates of the address are different between the form A and the form B, the LCD display unit 5 of the client terminal 51 displays an instruction message “Please tap the address” based on the instruction from the management server 61 side. (S15). In response to this guidance, for example, an area of coordinates (x4, y4) where the address on the image of the form 1 is located is designated (S16).

  The management server 61 refers to the form definition DB 22 and narrows down the form definition whose address is at the coordinates (x4, y4) (S17). As a result, the management server 61 (identification processing unit 37a) determines that the form for which the image has been acquired is the form B, and transmits the form definition corresponding to the form B (form B) to the client terminal 51. (S18). The client terminal 51 receives a form definition body corresponding to the transmitted form B (type of form B) (S19).

  As described above, according to the form identification system 41 according to the second embodiment, the number of times of information transmission / reception between the client terminal 51 and the management server 61 is reduced due to the efficiency of narrowing down the type of form. Thereby, the communication capacity between the client terminal 51 and the management server 61 is reduced, and the effort required for the designation operation for the user is also reduced.

<Third Embodiment>
Next, a third embodiment will be described with reference to FIGS. In FIG. 17, the same components as those in FIG. 5 described in the first embodiment are given the same reference numerals, and redundant descriptions are omitted. As illustrated in FIG. 17, the form identification system 42 according to the present embodiment includes a client terminal 52 instead of the client terminal 50 provided in the form identification system 40 of the first embodiment. The client terminal 52 includes a feature point extraction instruction unit 26b and a feature point information extraction unit 28b instead of the feature point coordinate specification unit 26 and the feature point information extraction unit 28 included in the client terminal 50 of the first embodiment. ing. The client terminal 52 further includes a feature point item DB (feature point item database) 39.

  In the first embodiment, the feature point coordinate designating unit 26 specifies the coordinates of a characteristic part in the image of the form as an element (item name) that defines the form type (style) of the feature point information every time it is extracted. However, in this embodiment, the designation operation from the user is not necessary. Upon receiving an instruction to extract feature point information from the feature point extraction instructing unit 26b, the feature point information extracting unit 28b, as shown in FIG. 18, has a plurality of elements (for example, a plurality of item names) defining the form type. ), Feature point information corresponding to a predetermined element (predefined item name) is automatically extracted. The feature point information extraction unit 28b has a character recognition unit 28a.

  The character recognizing unit 28a refers to a character recognition database (character recognition dictionary) in which character feature amounts (feature vectors and the like) and character codes (text data) are stored in association with each other. Character recognition is performed and the character code that is the character recognition result is output.

  That is, as shown in FIG. 19, in the feature point item DB (feature point item database) 39, item names corresponding to feature point information to be automatically extracted are stored in advance. The feature point information extraction unit 28b recognizes the image of the form 1 and further refers to the feature point item DB 39, and the items stored in the feature point item DB 39 in the image of the form 1 as shown in FIG. Feature point information in which all coordinate positions where names exist as character strings and their item names are associated is extracted.

  As described above, according to the form identification system 42 according to the third embodiment, when the feature point information is obtained, the designation operation such as the item name of the form is not necessary, which reduces the labor for the user. It is possible to shorten the time until the form definition is finally obtained.

<Fourth Embodiment>
Next, a fourth embodiment will be described with reference to FIGS. In FIG. 20, the same components as those in FIG. 5 described in the first embodiment are given the same reference numerals, and redundant descriptions are omitted. In FIG. 20, in order to clarify the structural explanation of the present embodiment, the feature point information transmitting unit 29, the form candidate list receiving unit 30, and the form ID on the client terminal 50 side shown in FIG. Transmission unit 31, form definition body reading unit 38, feature point information reception unit 33 on the management server 60 side, form candidate list transmission unit 34, form ID reception unit 35, form definition body transmission unit 36, form identification function unit 19, illustration of the form definition body reading function unit 20 is omitted.

  As illustrated in FIG. 20, the form identification system 43 according to the present embodiment replaces the client terminal 50 and the management server 60 included in the form identification system 40 according to the first embodiment with a client terminal 53 and a management server 63. Have As shown in FIG. 20, in addition to the configuration of the client terminal 50, the client terminal 53 further includes a form definition information transmission unit 31c that functions as a third transmission unit. In addition to the configuration of the management server 60, the management server 63 further includes a form definition body writing function unit 20c including a database construction unit 38c and a form definition information receiving unit 35c functioning as a third receiving unit.

  As shown in FIG. 21, the form identification system 43 has a registration mode for registering a form definition body as an operation mode for constructing the form definition body DB 22. That is, the form definition information transmitting unit 31c on the client terminal 53 side includes type identification information (for example, form ID) specific to the type of the form 1 from which the image has been acquired, and the feature point information extracted by the feature point information extracting unit 28. , And form definition information (form definition body) is transmitted.

  On the other hand, the form definition information receiving unit 35c on the management server 63 side receives the form definition information (form definition body) transmitted from the client terminal 53 side. The database construction unit 38c sequentially registers the received form definition information (form definition body) to construct the form definition body DB22. After the form definition DB 22 is constructed, in the normal identification mode of the form, the identification processing unit 37 is obtained from the image of the form to be identified while referring to the form definition DB 22 constructed in this way. Identification processing is performed based on the feature point information.

  Therefore, according to the form identification system 43 according to the fourth embodiment, it is possible to construct the form definition DB 22 without separately preparing an external system other than the form identification system.

<Fifth Embodiment>
Next, a fifth embodiment will be described with reference to FIG. The form identification system according to the present embodiment includes a client terminal 54 as shown in FIG. 22 instead of the client terminal 50 included in the form identification system 40 of the first embodiment. Since the image acquisition unit 63 of the client terminal 54 does not have the camera unit 4, the image acquisition unit 63 acquires the image of the form 1 from the outside of the client terminal 54 main body.

  That is, the storage 65 is provided outside the client terminal 54, the form 1 is imaged in advance by the digital camera 67, the scanner 64, and the like, and the obtained image of the form 1 is stored in the storage 65. Examples of the storage 65 include network storage such as a cloud service in addition to a hard disk device and a flash memory.

  When the client terminal 54 acquires the image of the form 1, the form image writing unit 23 accesses the external storage 65 through a network or a communication cable to acquire the image of the form 1 to be processed. It is written in the image DB 21.

  As described above, according to the form identification system of the fifth embodiment, the image of the form 1 can be obtained even at the client terminal 54 without the camera function. For example, a plurality of people share the work. It is possible to create an image of the form in advance.

<Sixth Embodiment>
Next, a sixth embodiment will be described based on FIG. 23 (and FIGS. 4, 5, and 10). In FIG. 23, the same components as those in FIG. 5 described in the first embodiment are denoted by the same reference numerals, and redundant description is omitted. As shown in FIG. 23, a form reading system 55 according to the present embodiment is replaced with a client terminal 55 and a management server 65 instead of the client terminal 50 and the management server 60 provided in the form reading system 40 of the first embodiment. have.

  The client terminal 55 further includes a form character recognition unit 73 in addition to the configuration of the client terminal 50. Further, the client terminal 55 has a communication function unit 17a instead of the communication function unit 17 provided in the client terminal 50. In addition to the configuration of the communication function unit 17, the communication function unit 17 a further includes an identification information receiving unit 72. On the other hand, the management server 65 further includes an identification information transmitting unit 71 in addition to the configuration of the management server 60. Further, the client terminal 55 stores the form definition TBL22c having the configuration illustrated in FIG. 10 in, for example, the storage unit 8 illustrated in FIG.

  Here, as shown in FIG. 23, the identification information transmitting unit 71 on the management server 65 side is the type identification information specific to the type of the form 1 obtained from the result of the identification processing by the identification processing unit 37 (see FIG. 5). (Form ID) is transmitted to the client terminal 55 side. The identification information receiving unit 72 on the client terminal 55 side receives the type identification information as the form type transmitted from the management server 65 side.

  As shown in FIG. 23, the form character recognition unit 73 sets the image of the form acquired by the image acquisition unit 13 as a recognition target image. That is, the form character recognition unit 73 recognizes the form image based on the type identification information received by the identification information reception unit 72. Specifically, the form character recognition unit 73 determines a form definition body (form definition information) corresponding to the received type identification information (form ID) while referring to the form definition body TBL22c illustrated in FIG. Then, character recognition is performed using the determined form definition body.

  As a result, the client terminal 55 side has the form definition body (each item name existing on the form, the coordinate position on the form to which the description area for each item name is assigned, and the form type specific to each item name). A data file in which the type identification information and the size of the form are linked to each other), and character recognition of the image part in the specified item name description area in the form image for each type of form Then, the character data (each text) that is the character recognition result can be divided into items and registered.

  Further, instead of the configuration described above, the identification processing unit 37 on the management server 65 side identifies the form definition information (form definition body) corresponding to the acquired form image (recognition target image), and this identified form The definition information may be transmitted from the identification information transmitting unit 71 on the management server 65 side to the identification information receiving unit 72 on the client terminal 55 side. In this case, the client terminal 55 can perform character recognition using the received form definition information directly without using the form definition body TBL22c shown in FIG.

  As mentioned above, although embodiment of this invention was described, these embodiment is shown as an example and is not intending limiting the range of invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

  DESCRIPTION OF SYMBOLS 1 ... Form, 2 ... Network, 13, 63 ... Image acquisition part, 22 ... Form definition body DB, 25, 25a ... Screen generation part, 26 ... Feature point coordinate designation part, 28, 28b ... Feature point information extraction part, 28a ... Character recognition unit, 29 ... Feature point information transmission unit, 30a ... Request information reception unit 31c ... Form definition information transmission unit, 33 ... Feature point information reception unit, 34a ... Request information transmission unit, 35c ... Form definition information reception unit , 37, 37a ... identification processing unit, 38c ... database construction unit, 40, 41, 42, 43, 45 ... form identification system, 50, 51, 52, 53, 54, 55 ... client terminal, 60, 61, 62, 63, 65 ... management server, 71 ... identification information transmission unit, 72 ... identification information reception unit, 73 ... form character recognition unit.

Claims (6)

A form identification system including a client terminal and a management server connected via a network,
The client terminal is
An image acquisition unit for acquiring a form image;
A feature point information extraction unit that extracts feature points of the form as feature point information from the acquired image;
A transmission unit for transmitting the extracted feature point information;
With
The management server
A receiving unit for receiving the transmitted feature point information;
Based on the received feature point information, an identification processing unit that performs identification processing of the type of the form,
A second transmission unit for transmitting request information for requesting designation of a predetermined element from among a plurality of elements defining the type of form;
Bei to give a,
The client terminal is
A second receiving unit for receiving the transmitted request information;
Based on the received request information, a screen generation unit that generates a guidance screen for guiding the designation of the predetermined element;
An element selection unit that selects an element designated and operated in accordance with the guidance from the plurality of elements;
Further comprising
The feature information extraction unit, from the image of the acquired form, extracts the feature point information corresponding to the selected element, the book form identification system. The identification process is a process of narrowing down the type of the form into several types, or a process of specifying the type of the form as one type.
The form identification system according to claim 1. The feature point information extraction unit extracts feature point information corresponding to a predetermined element from a plurality of elements defining the type of form,
The form identification system according to claim 1 or 2. The client terminal is
A third transmission unit for transmitting form definition information in which the type identification information unique to the type of the form from which the image has been acquired and the extracted feature point information are associated;
The management server
A third receiving unit for receiving the transmitted form definition information;
A database construction unit that constructs a database by registering the received form definition information;
Further comprising
The identification processing unit performs the identification processing while referring to the constructed database.
The form identification system according to claim 1 or 2. The image acquisition unit acquires the image of the form from outside the client terminal body;
The form identification system according to any one of claims 1 to 4 . The management server
An identification information transmission unit that transmits type identification information specific to the type of the form obtained from the result of identification processing by the identification processing unit to the client terminal;
The client terminal is
An identification information receiving unit for receiving the type identification information transmitted from the management server;
A form character recognition unit that uses the image of the form acquired by the image acquisition unit as a recognition target image;
Further comprising
The form identification system according to any one of claims 1 to 5 , wherein the form character recognition unit recognizes characters of the form image based on the type identification information received by the identification information reception unit.

Images