Document reading device and document reading program

Abstract

Provided is a document reading device comprising: a mounting table on which a document and an index board are placed; an imaging unit that captures an image of the mounting table and outputs an entire image of the mounting table; a recognition unit that recognizes, from the entire image, a document region of the document and the index board region of the index board placed on the mounting table; and a processing unit that executes a setting process related to reading a document region preset for the index board, on the basis of at least one of the shape, position, and orientation of the index board region recognized by the recognition unit. By configuring the document reading device in this manner, it is possible to provide a user interface that can make effective use of a wide mounting table and sufficiently improve work efficiency.

Description

Document reading device and document reading program

[0001] The present invention relates to a document reading device and a document reading program.

[0002] Scanners that scan paper documents using a line sensor and convert the scanned data into digital files are now widely used. A typical scanner scans a document placed in a fixed position on a transparent glass document tray. A document camera-type document reader, which is a further development of such scanners as a copy machine application, has become known. It uses an image sensor with two-dimensionally arranged pixels to capture an overhead image of the entire document placed on the tray, thereby scanning the entire document at once (see, for example, Patent Document 1).

[0003] Japanese Patent Application Laid-Open No. 2018-207342

[0004] Previous document camera-type document reading devices were limited by the limitations of their image sensors, and could only hold a single document on their platform. However, with improvements in image sensor performance, it has become possible to obtain sufficient resolution even when capturing images of a large platform at once. This has led to the development of document reading devices with platforms that can hold multiple A4-sized documents, which are commonly used.

[0005] A document reading device with such a wide platform is highly convenient because it can read multiple documents side by side at once and instantly generate a read file corresponding to each document. For example, if three application documents are accepted for each applicant, by scanning all three application documents at once, a read file can be generated for each applicant, and these documents can also be managed as a group.

[0006] In such document reading devices, the user must repeatedly arrange and remove documents to be read on the table, and operate an operation panel, mouse, etc. to instruct the reading process. It can be said that conventional user interfaces that use an operation panel, mouse, etc. do not fully utilize the convenience of document reading devices with a large table.

[0007] The present invention has been made to solve such problems, and provides a document reading device, etc., equipped with a user interface that can effectively utilize a wide mounting base to sufficiently improve workability.

[0008] A document reading device in a first aspect of the present invention comprises a placing table on which a document and an index board are placed, an imaging unit that images the placing table and outputs an overall image of the placing table, a recognition unit that recognizes the document area of ​​the document placed on the placing table and the index board area of ​​the index board from the overall image, and a processing unit that executes setting processing for reading a document area that has been preset for the index board based on at least one of the shape, position, and orientation of the index board area recognized by the recognition unit.

[0009] In addition, a document reading program in a second aspect of the present invention is a document reading program for a document reading device that includes a table on which a document and an index board are placed, and an imaging unit that images the table and outputs an overall image of the table, and causes a computer to execute a recognition step that recognizes the document area of ​​the document placed on the table and the index board area of ​​the index board from the overall image, and a processing step that executes a setting process for reading a document area that has been preset for the index board based on at least one of the shape, position, and orientation of the index board area recognized in the recognition step.

[0010] The present invention can provide a document reading device or the like equipped with a user interface that can effectively utilize a wide table and sufficiently improve workability.

[0011] 1 is a diagram illustrating an example of a usage status of a document reading device according to an embodiment; FIG. 1 is a diagram illustrating a hardware configuration of a document reading device; FIG. 2 is a diagram illustrating a usage mode of an index board and a set process; FIG. 3 is a diagram illustrating a usage mode of another index board and a set process; FIG. 4 is a diagram illustrating a usage mode of another index board and a set process; FIG. 5 is a diagram illustrating a usage mode of another index board and a set process; FIG. 6 is a diagram illustrating a usage mode of another index board and a set process; FIG. 7 is a diagram illustrating a usage mode of another index board and a set process; FIG. 8 is a diagram illustrating a usage mode of another index board and a set process; FIG. 9 is a diagram illustrating a usage mode of another index board and a set process; FIG. 10 is a diagram illustrating a usage mode of another index board and a set process; FIG. 11 is a diagram illustrating a usage mode of another index board and a set process; FIG. 12 is a diagram illustrating a usage mode of another index board and a set process; FIG. 13 is a diagram illustrating a usage mode of another index board and a set process; 10A and 10B are diagrams for explaining the use modes of other indicator boards and the set processes, and FIG. 10A is a diagram for explaining the use modes of other indicator boards and the set processes, and FIG. 10B is a diagram for showing a pre-setting screen for the indicator board and the setting processes,

[0012] The present invention will be described below through embodiments of the invention, but the invention according to the claims is not limited to the following embodiments. Furthermore, not all of the configurations described in the embodiments are necessarily essential as means for solving the problems.

[0013] 1 is a diagram showing an example of a usage situation of a document reading device 100 according to this embodiment. The document reading device 100 is a device that can capture images of multiple documents at once and generate read files corresponding to each document from the captured images.

[0014] The document reading device 100 mainly comprises a mounting table 101, a control unit 110, an imaging unit 120, an illumination unit 130, a display unit 140, and an input unit 150. The mounting table 101 is a so-called document table that is large enough to accommodate, for example, three A4-sized documents arranged two by three, and is used to place multiple documents on the mounting table 101. In addition to the documents to be read, an index plate that causes the document reading device 100 to perform processes related to document reading is also placed on the mounting table 101. For example, a low-reflectivity black sheet is attached to the surface of the mounting table 101.

[0015] In the illustrated example, three documents to be read (document A 901, document B 902, and document C 903) and two round index boards 800 are placed on the table 101. As will be described in detail later, when a user covers the two round index boards 800, which are placed in positions that do not overlap the documents, with their right and left hands, respectively, the document reading device 100 generates read files corresponding to each of the three documents. In this embodiment, various index boards other than the index board 800 are prepared and are placed near the table 101 so that the user can place them on the table 101 as appropriate. The figure shows the state in which some of the index boards, such as index board 815, are prepared.

[0016] The control unit 110 houses a control element, a control circuit, etc. for controlling the document reading device 100. The control unit 110 may be a general-purpose computer such as a PC, and when the control unit 110 is a PC, the entire unit functions as the document reading device 100 by executing a document reading program.

[0017] The imaging unit 120 is installed so as to overlook the mounting table 101 and includes an imaging lens and an imaging sensor capable of capturing an image of the entire mounting table 101 at a certain resolution. The imaging unit 120 captures an image of the mounting table 101 in accordance with a request from the control unit 110 and transmits the generated image signal to the control unit 110. The imaging unit 120 may be configured using, for example, a general-purpose mirrorless single-lens camera rather than a dedicated imaging unit. The illumination unit 130 uniformly illuminates the entire mounting table 101. The illumination unit 130 is installed so as to overlook the mounting table 101 in the same manner as the imaging unit 120 and includes, for example, a white LED and a diffusion plate that diffuses the emitted light. The illumination unit 130 illuminates the mounting table 101 in accordance with instructions from the control unit 110. The illumination unit 130 may be configured using a projector that can project various information onto the mounting table 101.

[0018] The display unit 140 is, for example, a liquid crystal monitor, and is installed in a position that allows easy viewing by a user who places or removes a document to be scanned on the document table 101. The display unit 140 receives a display signal from the control unit 110 and displays an overall image of the document table 101 captured by the imaging unit 120. The figure shows a display of an overall image including documents A 901, B 902, and C 903 placed on the document table 101 and two round indicator boards 800. The input unit 150 is an input device provided separately from the user interface using the indicator boards described below, and allows the user to give instructions to the control unit 110. The input unit 150 may also include a touch panel superimposed on the display surface of the display unit 140, a microphone linked to voice instruction recognition, or the like.

[0019] 2 is a hardware configuration diagram of the document reading device 100. The control unit 110 mainly includes a control unit 160, a storage unit 170, and an image processing unit 180. The control unit 160 is a processor (CPU: Central Processing Unit) that controls the document reading device 100 and executes programs. The processor may be configured to work in conjunction with a processing chip such as an ASIC (Application Specific Integrated Circuit) or a GPU (Graphics Processing Unit). The control unit 160 reads out a document reading program stored in the storage unit 170 and executes various processes related to document reading.

[0020] The storage unit 170 is a non-volatile storage medium, such as a solid state drive (SSD). In addition to storing the generated read file, the storage unit 170 also stores constants, variables, setting values, CG data, document reading programs, and the like required for operation of the document reading device 100. The image processing unit 180 is, for example, configured by an ASIC that exclusively performs image processing, and performs various image processing such as interpolation processing on the image signal received from the imaging unit 120 to generate image data conforming to a predetermined format and pass it to the control unit 160.

[0021] The imaging unit 120, illumination unit 130, display unit 140, and input unit 150 are connected to the control unit 160 by wire or wirelessly. The imaging unit 120 performs imaging processing in response to an imaging request from the control unit 160 and returns the output imaging signal to the control unit 160. For example, when an imaging signal for display is requested, the imaging unit 120 repeats the imaging processing at a cycle of 60 fps and sequentially transmits the resulting frame signals to the control unit 160 as imaging signals. In this case, the image processing unit 180 generates 60 fps image data from the frame signals delivered from the control unit 160, and the image data is converted by the control unit 160 into a display signal and displayed on the display unit 140 as a 60 fps display image.

[0022] The illumination unit 130 causes the LED to emit light in response to an illumination instruction from the control unit 160, and stops the LED from emitting light in response to a stop signal. The display unit 140 receives the display signal generated and converted by the control unit 160, converts it into a visible monitor image, and displays it. The input unit 150 converts a user's operation into an operation signal and transmits it to the control unit 160.

[0023] The control unit 160 plays various roles as a function control unit depending on the document reading program being executed. For example, it functions as a display control unit when generating a display signal for a monitor image to be displayed on the display unit 140. In this embodiment, the control unit 160 can particularly function as a recognition unit 161, a reception unit 162, and a processing unit 163. The recognition unit 161 mainly recognizes the document areas of each of the multiple documents placed on the table 101 and the index board area of ​​the index board also placed on the table 101 from the overall image generated by the image processing unit. The reception unit 162 mainly cooperates with the input unit 150 to receive from the user settings for the index board and operations for the read file after reading. The processing unit 163 mainly executes a preset setting process for the index board based on the position of the index board area recognized by the recognition unit 161, and cooperates with the image processing unit 180 to cut out areas corresponding to the frame indicators from the overall image and generate multiple read files.

[0024] Document reading devices with a wide platform, such as the document reading device 100 of this embodiment, are generally highly convenient because they can arrange and read multiple documents at once and instantly generate scanned files corresponding to each document. For example, if three application documents are accepted per applicant, scanning all three documents at once allows scanned files to be generated for each applicant, and these documents can also be managed as a group. Even with such a document reading device, if a conventional user interface using an operation panel, mouse, or the like was adopted, it was necessary to repeatedly arrange and remove documents to be read on the platform and to perform operations related to the scanning process.

[0025] The document reading device 100 of this embodiment employs a user interface that utilizes index plates to eliminate such cumbersome tasks and fully utilize the convenience of a document reading device equipped with a large table. The index plates are plates cut into various shapes, most measuring approximately 5 cm x 5 cm. The plates are made of materials such as plastic or aluminum. Individual settings related to document area reading are pre-configured for each index plate, and the settings are compiled into a settings file and stored in the storage unit 170. When a user selects an index plate corresponding to the desired process and places it in an appropriate position on the table 101, the processing unit 163 recognizes the index plate from the acquired image and executes the pre-configured process based on its position, orientation, etc. With this user interface using index plates, the user does not need to operate operating members each time to generate a scanned file. Instead, they only need to focus on the work performed on the table 101. This effectively utilizes the large table and significantly improves work efficiency.

[0026] Below, we will explain how to use several index boards and the setting processes set for each index board, using specific examples of display screens, etc. Figure 3 is a diagram for explaining how to use the index board 800 and the processes set for the index board 800. Specifically, the upper diagram shows an example of a display screen that displays an entire image of the mounting base 101 in the situation of Figure 1 where two index boards 800 are placed, and the lower diagram shows the structure of the read file generated in response to a file generation instruction from the user.

[0027] The display unit 140 displays an overall image 200 of the document placement platform 101 together with several CG images. The overall image 200 includes images of documents A901, B902, and C903, each of which is A4 size. The images are automatically recognized by the recognition unit 161 as document areas A201, B202, and C203, respectively. Recognition of each document area from the overall image 200 can be performed, for example, by converting the overall image to a black-and-white image using a predetermined threshold value and matching the black-and-white boundary with a rectangular shape. By employing such automatic recognition, the document area can be properly recognized even if the document is placed at an angle relative to the document placement platform 101. In other words, the user can quickly place the document for which a scanned file is to be generated on the document placement platform 101 without worrying about its orientation.

[0028] A frame index A211, a frame index B212, and a frame index C213, which are border frames of a fixed width, are superimposed on the outer edges of the recognized document areas A201, B202, and C203, respectively. Each of the index frames is CG and is distinguished from one another, for example, by color. While a border frame is used in this embodiment, any other method may be used as long as the document area can be visually recognized by the user, such as placing L-shaped pointers at the four corners.

[0029] Identification indicators A221, B222, and C223 indicating identification symbols for identifying document areas are superimposed near frame indicator A211, frame indicator B212, and frame indicator C213, respectively. In this embodiment, the identification symbols are sequential numbers such as "1," "2," and "3" to indicate the order of the document areas included in the displayed entire image 200. The identification symbols displayed as identification indicators are not limited to sequential numbers, and may be part of the file name assigned when the scanned file is generated, or may be based on the document content recognized by the OCR function.

[0030] 1, the two index plates 800 are placed at positions on the table 101 so as not to overlap with documents A901, B902, and C903. The user can place the two index plates 800 at any positions on the table 101 after or before placing documents A901, B902, and C903.

[0031] The entire image 200 includes images of these two index boards 800, each of which is automatically recognized by the recognition unit 161 as an index board area 300. As with document area recognition, the recognition unit 161 extracts the outline of the index board area 300 and performs pattern matching with the pre-set shapes of each index board to recognize the shape of the index board area 300 and its position in the coordinate system of the entire image 200. The illustrated index board area 300 is circular, so its orientation is not defined. However, depending on the index board area to be recognized, the recognition unit 161 also recognizes the orientation of the index board area relative to the coordinate system of the entire image 200 based on the shape of the index board area in order to recognize the orientation of the index board to be recognized placed on the placement table 101. Conversely, depending on the index board area to be recognized, it may be possible to recognize only the orientation without recognizing the position.

[0032] Furthermore, depending on the indicator board area to be recognized, the recognition unit 161 recognizes at least one of the color, pattern, and text information of the indicator board area. For example, even if indicator boards have the same shape, different processing may be set depending on the coloring of the surface. When the recognition unit 161 recognizes a shape corresponding to such an indicator board, it also recognizes the color of the indicator board area. The same applies to the pattern. In addition, there may be indicator boards for which different processing is set depending on the text information written on the surface. When the recognition unit 161 recognizes a shape corresponding to such an indicator board, it recognizes the text information written in the indicator board area using the OCR function.

[0033] Returning to the generation of the scanned file, first, the user can visually check the entire image 200 and frame indicators (frame indicator A211, frame indicator B212, frame indicator C213) displayed on the display unit 140 before the scanned file is generated using this display screen. Therefore, the user can intuitively imagine the scanned file to be generated from the document area (document area A201, document area B202, document area C203) surrounded by the frame indicators, and can quickly confirm whether it is what they intended. Then, once the user decides to generate a scanned file from the document area surrounded by the indicator frames, the user covers the two indicator plates 800 with their hands. For example, the user covers one of the indicator plates 800 located on the right side of the table 101 with their right hand and the other located on the left side with their left hand.

[0034] The circular index boards 800 are preset with a setting process that generates an instruction to generate a read file when both indexes are covered at the same time. That is, the user can issue an instruction to generate a read file to the processing unit 163 by making a hand gesture to "cover" the two index boards 800 placed on the mounting base 101. The recognition unit 161 monitors changes in the two recognized index board areas 300, and when it recognizes that the two index board areas 300 are covered by hands at the same time, it passes the recognition result to the processing unit 163. When the processing unit 163 receives the recognition result, it executes the read file generation process, assuming that the user has instructed it to generate a read file.

[0035] In this way, the user can issue an instruction to create a scanned file with a simple gesture of covering the index board 800 placed in a position that does not interfere with document scanning, thereby freeing the user from the hassle of operating operating members. Also, since the file creation instruction is initiated only by the combination of hand gestures of simultaneously covering two index boards 800, even if the user unintentionally covers one of the index boards 800 while working on the table 101, the file creation instruction is not initiated, and the index boards 800 are unlikely to interfere with the user's work.

[0036] When the processing unit 163 starts generating a read file based on the fact that the two index plates 800 are covered, it cooperates with the image processing unit 180 to cut out the areas corresponding to the frame indexes from the entire image 200 and generate the read file. Note that the image data of the entire image 200 to be cut out at this time may be the image data used for display on the display unit 140, or may be image data acquired by having the imaging unit 120 capture the image again. When having the imaging unit 120 capture the image again, the imaging may be performed in a high-resolution mode so that image data with a higher resolution than the image data to be displayed on the display unit 140 can be acquired.

[0037] As shown in the figure, the processing unit 163 generates a read file "docfile_01.pdf" from the document area A201 surrounded by the frame index A211, a read file "docfile_02.pdf" from the document area B202 surrounded by the frame index B212, and a read file "docfile_03.pdf" from the document area C203 surrounded by the frame index C213. The sequential numbers "01," "02," and "03" in the file names correspond to the sequential numbers of the identification indexes. In other words, the order of the generated read files is determined according to the order indicated by or contained within the identification symbols of the identification indexes. Note that, although the generated read files are PDF files in this embodiment, the file format of the generated read files is not limited to PDF files.

[0038] The processing unit 163 performs an association process to associate multiple scanned files generated based on a single file creation instruction. That is, some sort of association process is performed on the generated scanned files so that the multiple scanned files generated based on a single file creation instruction can be recognized as related files. In this embodiment, a folder is generated to store the multiple scanned files generated based on a single file creation instruction. Specifically, in the tree structure of the storage unit 170, a folder with a consecutive number "docfol_05" is generated at a set level, and the three scanned files generated are stored in the folder. The association process is not limited to folder generation. For example, the ID of the scanned file generated in conjunction with the scanned file may be written in the tag area of ​​each scanned file. Furthermore, the generated scanned files may be combined into a single document file, each of which represents one page according to its order.

[0039] 4A and 4B are diagrams for explaining the usage of the index plate 801 and the processing set for the index plate 801. In particular, Fig. 4A is a diagram showing the state in which the index plate 801 is placed on the placement table 101, and Fig. 4B is a diagram showing the state before and after the index plate area 301, which is the image of the index plate 801, appears in the entire image 200.

[0040] The index plate 801 is L-shaped and is used to change the shape of a recognized document area. When a user wants to change the shape of a specific document area, the user places the index plate 801 on top of or near the target document. Here, we will explain how to change the shape of document area A201 when reading document A901.

[0041] As shown in FIG. 4A, the user places the index plate 801 over the area of ​​the document A901 placed on the table 101 for which a scanned file is to be generated. Then, as shown in the left diagram of FIG. 4B, the frame index A211, which had previously surrounded the document area A211 corresponding to the entire document A901, is deformed to match the position and shape of the index plate area 301, which is the image of the index plate 801, as shown in the right diagram of FIG. 4B. Specifically, the common portion of the index plate area 301 between the narrow-angle sides of the two L-shaped sides and the original document area A201 is determined as the new document area A201, and the frame index A211 is deformed to surround this new document area A201. In other words, the processing unit 163 changes the document area A201 to the area of ​​the image of the document A901 that is surrounded by the deformed frame index A211. When an instruction to generate a read file is subsequently received from the user, the processing unit 163 cuts out a new document area A201 and generates a read file.

[0042] Figure 5 is a diagram for explaining the usage mode when index plate 801 and optional index plate 802 are used together, and the processing set for index plate 802. Figure 5(A) is a diagram showing the state when index plate 801 and index plate 802 are placed on mounting table 101, and Figure 5(B) is a diagram showing the state after index plate area 301, which is the image of index plate 801, and index plate area 302, which is the image of index plate 802, have appeared in the overall image 200.

[0043] The indicator plate 802 has a small circular shape and is colored the same as the L-shaped indicator plate 801 that is used together with the indicator plate 802. The indicator plate 802 is used together with the indicator plate 801 to change the shape of the recognized document area. The indicator plate 802 is colored differently from the indicator plate 800, and the recognition unit 161 recognizes the two plates separately.

[0044] As shown in Fig. 5A, the user places index boards 801 and 802 on document A901 placed on the table 101, matching them to the area for which a scanned file is to be generated. Then, as shown in Fig. 5B, frame index A211 is deformed to match the position and shape of index board area 301, which is the image of index board 801, and the position of index board area 302, which is the image of index board 802. Specifically, for index board area 301, an area sandwiched between the narrow-angle sides of two sides forming an L-shape is defined as described with reference to Fig. 4, and for index board area 302, two sides that pass through that position and form a rectangle with the two sides formed by index board area 301 are defined. In this way, the common portion between the rectangular area defined by the two index board areas and the original document area A201 is determined as a new document area A201, and frame index A211 is deformed to surround this new document area A201. That is, the processing unit 163 changes the document area A201 to an area surrounded by the deformed frame index A211 in the image of the document A901. If an instruction to generate a read file is subsequently received from the user, the processing unit 163 cuts out the new document area A201 and generates a read file.

[0045] 6A and 6B are diagrams for explaining a usage mode when index plates 803 and 804 are used together, and a process set for the combination of index plates 802 and 804. In particular, Fig. 6A is a diagram showing the state in which index plates 803 and 804 are placed on table 101, and Fig. 6B is a diagram showing the state after index plate region 303, which is the image of index plate 803, and index plate region 304, which is the image of index plate 804, have appeared in overall image 200.

[0046] The index plate 803 has the same shape as the index plate 801, but is colored differently from the index plate 801. Similarly, the index plate 804 has the same shape as the index plate 802, but is colored differently from the index plate 802 and is colored the same as the index plate 803. The index plates 803 and 804 are used to change the shape of the recognized document area. The index plate 804 is also colored differently from the index plate 800, and the recognition unit 161 recognizes them separately.

[0047] As shown in Fig. 6(A), the user places indicator boards 801 and 802 in areas of document A901 placed on table 101 where the user does not want to generate a scanned file. Then, as shown in Fig. 6(B), frame indicator A211 is transformed into two frame indicators 211a and 211b that are divided according to the position and shape of indicator board area 303, which is the image of indicator board 803, and the position of indicator board area 304, which is the image of index board 804. Specifically, a rectangular area similar to the rectangular area defined by indicator board area 301 and indicator board area 302 in the example of Fig. 5 is defined as exclusion area 201c by indicator board area 303 and index board area 304. Then, document area 201a and document area 201b obtained by excluding exclusion area 201c from original document area A201 are determined as new document areas, and frame index 211a is superimposed so as to surround document area 201a, and frame index 211b is superimposed so as to surround document area 201b.

[0048] The processing unit 163 determines the document area 201a and the document area 201b as new document areas for the document A 901. Then, when an instruction to generate a read file is received from the user, the processing unit 163 cuts out the document area 201a and the document area 201b from the entire image 200 and connects them vertically to generate one read file.

[0049] In the combination of index board 803 and index board 804, the process of cutting out the corresponding document areas and connecting them vertically has been set, but the exclusion process is not limited to this. For example, in the combination of index boards with the same shape but different colors, a setting process may be adopted in which the original document area A201 is cut out and a read file is generated in which the area corresponding to excluded area 201c is blacked out, or in which document area 201a and document area 201b are cut out and two read files are generated.

[0050] 7A and 7B are diagrams for explaining a usage mode when index plates 805 and 806 are used together, and a process set for the combination of index plates 805 and 806. In particular, Fig. 7A is a diagram showing the state in which index plates 805 and 806 are placed on table 101, and Fig. 7B is a diagram showing the state after index plate region 305, which is the image of index plate 805, and index plate region 306, which is the image of index plate 806, have appeared in overall image 200.

[0051] Indicator plate 805 is shaped like an arrow and has the character "left" written on its surface. Indicator plate 806 is shaped like an arrow and has the character "right" written on its surface, similar to indicator plate 805. Indicator plate 805 and indicator plate 806 are used to instruct the placement of multiple recognized document areas and generate a single combined document area.

[0052] As shown in FIG. 7A, the user places index board 805 with the arrow pointing toward document A901 placed on table 101. Also, the user places index board 806 with the arrow pointing toward document B902, which is placed on table 101 to the left of document A901. Then, as shown in FIG. 7B, processing unit 163 generates a new document area 201d by arranging original document area A201, which is an image of document A901, on the right and original document area B202, which is an image of document B902, on the left, in accordance with the instructions of the characters written on each index board. Since document area 201d is not actually included in overall image 200, a new area window 401 is provided to be superimposed on overall image 200, and document area 201d, which is obtained by cutting out document area A201 and document area B202 from overall image 200 and arranging them side by side, is displayed in new area window 401. Then, a frame indicator 211d is displayed so as to surround the document area 201d.

[0053] The processing unit 163 determines this document area 201d as a new document area replacing document area A201 and document area B202. Then, upon receiving a user instruction to generate a scanned file, the processing unit 163 cuts out document area A201 and document area B201 from the entire image 200 and connects them horizontally to generate a single scanned file. While the example in FIG. 7 illustrates the combination of index board 805 with the character "left" and index board 806 with the character "right," the combination of index boards is not limited to this. For example, if the combination of index boards includes the characters "up" and "down," respectively, a process can be set to generate a new document area by arranging the respective document areas vertically and combining them. Furthermore, the document areas to be combined do not need to be the same size. For example, when combining document areas corresponding to A4 size and A5 size, the document areas are arranged according to a specified layout, and the rectangular area containing them is defined as the combined document area.

[0054] 8A and 8B are diagrams for explaining the usage of the index plate 807 and the processing set for the index plate 807. In particular, Fig. 8A is a diagram showing the state in which the index plate 807 is placed on the placement table 101, and Fig. 8B is a diagram showing the state after the index plate area 307, which is the image of the index plate 807, appears in the entire image 200.

[0055] The index plate 807 is hexagonal and has the letters "A4" written on its surface. The index plate 807 is used to enlarge or reduce all of the recognized document areas to the document size written on it, regardless of the actual size of the document, and generate a read file.

[0056] As shown in Fig. 8A, the user places index plate 807 in an empty space on table 101 on which A4-sized document A901, A3-sized document B902, and A5-sized document C903 have been placed. Then, as shown in Fig. 7B, character information of the document size originally recognized by recognition unit 161 and character information of the document size specified by index plate 807 are displayed via arrows next to the identification indices indicating each document area. At this time, document area A201, document area B202, and document area C203 recognized by recognition unit 161, and the frame indices A211, B212, and C213 surrounding them, respectively, remain unchanged.

[0057] When the processing unit 163 receives an instruction from a user to generate a read file, it generates an A4-sized read file for document area A201, which is the same size as the original document A901, generates a read file for document area B202 by reducing the A3 size of the original document B902 to A4 size, and generates a read file for document area C203 by enlarging the A5 size of the original document C903 to A4 size.

[0058] 9A and 9B are diagrams for explaining the use of the index plate 808 and the processing set for the index plate 808. In particular, Fig. 9A is a diagram showing the state in which the index plate 808 is placed on the placement table 101, and Fig. 9B is a diagram showing the state after the index plate area 308, which is the image of the index plate 808, has appeared in the entire image 200.

[0059] The index plate 808 has a circular ring shape. The index plate 808 is used to select a document for which a scanned file is to be generated from among multiple recognized document areas. As shown in FIG. 9A , the user places the index plate 808 near documents A901 and B902, which are documents for which scanned files are to be generated, among documents A901, B902, and C903 placed on the document table 101. Then, as shown in FIG. 9B , the frame index C213 is removed from document area C203, which does not have the index plate area 308, which is the image of the index plate 808, in its vicinity. At this time, the frame indexes A211 and B212 surrounding document area A201 and document area B202, respectively, do not change.

[0060] When the processing unit 163 receives an instruction from the user to generate a read file, it extracts each document area A201 and document area B202 from the entire image 200 and generates a read file, but does not generate a read file for document area C203.

[0061] 10A and 10B are diagrams for explaining the use of the index board 809 and the processing set for the index board 809. In particular, Fig. 10A is a diagram showing the state in which the index board 809 is placed near the document A 901 placed on the table 101, and Fig. 10B is a diagram showing the state after the index board area 309, which is the image of the index board 809, has appeared in the entire image 200.

[0062] The indicator plate 809 has a pentagonal shape. The indicator plate 809 is used to remove blank areas from the surrounding document area and update its shape. As shown in FIG. 10A, a user places the indicator plate 809 near a document A901 placed on the table 101. Here, document A901 is a document in which a driver's license is copied onto a regular A4-sized sheet of paper, and the portion where the driver's license appears is a portion of the entire sheet, with the rest being blank.

[0063] 10B, the frame index A211 that surrounded the document area A201 corresponding to the entire document A901 is deformed so that the margin area of ​​the document area A211 is outside the frame. That is, the processing unit 163 defines the rectangular area remaining after excluding the margin area of ​​the document area A201 as the new document area A201, and deforms the frame index A211 so that it surrounds this new document area A201. In this case, for example, a driver's license may be copied rotated relative to the document A901, and the rectangular area corresponding to the portion where the driver's license appears is determined as the new document area A201. When the processing unit 163 receives an instruction to generate a read file from the user, it cuts out the new document area A201 from the entire image 200 and generates a read file.

[0064] 11A and 11B are diagrams for explaining the use of the index board 810 and the processing set for the index board 810. In particular, Fig. 11A is a diagram showing the state in which the index board 810 is placed with the arrow direction pointing toward a specific description frame in the document A901 placed on the table 101, and Fig. 11B is a diagram showing the state after the index board area 310, which is the image of the index board 810, appears in the entire image 200.

[0065] The indicator board 810 has a shape combining an arrow and a circle. The indicator board 810 is used to search for a specific description frame located in the direction of the arrow in a nearby document and convert the description within that frame into text information using the OCR function. As shown in FIG. 11A , a user places the indicator board 810 with the arrow pointing toward a specific description frame in document A901 placed on the table 101 that they wish to convert into text information. Then, as shown in FIG. 11B , an OCR frame indicator 511 appears surrounding an OCR area 501, which is an area in document area A201 that corresponds to the specific description frame in document A901. That is, the processing unit 163 searches for a frame area corresponding to the specific description frame located in the direction of the indicator board 810 in document area A201, and designates the found frame area as the OCR area 501. The OCR frame indicator 511 is then generated and superimposed to surround the OCR area 501.

[0066] Furthermore, the processing unit 163 recognizes the character information written in the OCR frame area 501 using the OCR function. The recognized character information may be displayed near the OCR frame indicator 511. Note that the document area A201 recognized by the recognition unit 161 and the frame indicator A211 surrounding it do not change particularly when the indicator plate 810 is placed on the mounting table 101. When the processing unit 163 receives an instruction to generate a read file from the user, it cuts out the document area A201 from the entire image 200 and generates a read file. At this time, the character information recognized by the OCR function is associated with the read file as accompanying information. For example, the processing unit 163 embeds character data of the recognized character information in the area corresponding to the OCR area 501 of the read file instead of image data.

[0067] 12A and 12B are diagrams for explaining the usage of the index board 811 and the processing set for the index board 811. In particular, Fig. 12A is a diagram showing the state in which the index board 811 is placed on the document A901 placed on the table 101 with the arrow direction facing the document A901, and Fig. 12B is a diagram showing the state after the index board area 311, which is the image of the index board 811, appears in the entire image 200.

[0068] The indicator board 811 has a leaf-like arrow shape. The indicator board 811 is used to extract and indicate a specific description frame in which a pre-set inspection item is described from a document to which the arrow is directed. As shown in FIG. 12A , the user places the indicator board 811 with the arrow pointing toward document A901, which is the document to be inspected among the documents placed on the table 101. Then, as shown in FIG. 12B , inspection item frames 512a and 512b appear to surround the area in document area A201 corresponding to the specific description frame in which the specified inspection item is described. Furthermore, an inspection item indicator 522a indicating the specified inspection item "Name" appears near the inspection item frame 512a, and an inspection item indicator 522b indicating the specified inspection item "Address" appears near the inspection item frame 512b.

[0069] That is, the processing unit 163 extracts from the document area A201 areas corresponding to the specific description frames in which "name" and "address," which are inspection items preset for the index board 811, are written. Specifically, coordinate values ​​of the specific description frames corresponding to each inspection item are registered in the setting file, and the processing unit 163 extracts the corresponding areas from the document area A201 based on those coordinate values. Then, the processing unit 163 generates and superimposes inspection item frames 512a and 512b so as to surround each area. Furthermore, the processing unit 163 generates inspection item indicators 522a and 522b and places them near the inspection item frames 512a and 512b, respectively. When such a display is executed, the user can easily check whether the inspection items, name and address, are correctly written in the document to be read before generating the read file.

[0070] 13A and 13B are diagrams for explaining the usage of the index plate 812 and the processing set for the index plate 812. In particular, Fig. 12A is a diagram showing the state in which the index plate 812 is placed on the placement table 101, and Fig. 13B is a diagram showing the state after the index plate area 312, which is the image of the index plate 812, appears in the entire image 200.

[0071] The indicator plate 812 is star-shaped. The indicator plate 812 is used to check whether a specified character string is included in each of the recognized document areas. As shown in FIG. 13A , the user places the indicator plate 812 in an empty space on the table 101 on which document A 901, stacked documents B 902 a, 902 b, and 902 c, and document C 903 are placed. Then, as shown in FIG. 13B , a character string detection frame 513 appears in the document areas corresponding to each document, surrounding the character string area in which the previously specified character string "No. 02162" is detected.

[0072] That is, the processing unit 163 uses the OCR function to detect, from each document area, the character string "No. 02162," which is preset for the index plate 812. Then, the processing unit 163 generates and superimposes a character string detection frame 513 so as to surround each detected character area. When this display is executed, the user can easily check, before generating a scanned file, whether, for example, the same document ID is printed on multiple documents to be scanned. Note that the character string does not have to be a number; it could be, for example, the name of a specific organization or the date the document was created.

[0073] 14A and 14B are diagrams for explaining the usage of the index plate 813 and the processing set for the index plate 813. In particular, Fig. 14A is a diagram showing the state in which the index plate 813 is placed on the placement table 101, and Fig. 14B is a diagram showing the state after the index plate area 313, which is the image of the index plate 813, has appeared in the entire image 200.

[0074] The indicator plate 813 has a multi-projection shape. The indicator plate 813 is used to confirm whether a specified logo is included in each of the recognized document areas. As shown in FIG. 14A, the user places the indicator plate 813 in an empty space on the table 101 on which documents A 901, B 902, and C 903 are placed. Then, as shown in FIG. 14B, in the document areas corresponding to the documents, a logo detection frame 514 appears surrounding the logo area in which a logo 911, a pre-specified logo, is detected.

[0075] That is, the processing unit 163 uses pattern matching to detect the logo 911, which is a logo that has been set in advance for the indicator board 813, from each document area. Then, the processing unit 163 generates and superimposes a logo detection frame 514 so as to surround each detected logo area. This display allows the user to easily check, before generating a scanned file, whether multiple documents to be scanned contain, for example, an organizational logo representing the same publisher. The specified logo is not limited to a stylized combination of characters, but may be any mark indicating affiliation, identification, etc.

[0076] 15A and 15B are diagrams for explaining the use of the index plate 814 and the processing set for the index plate 814. In particular, Fig. 15A is a diagram showing the state in which the index plate 814 is placed on the placement table 101, and Fig. 15B is a diagram showing the state after the index plate area 314, which is the image of the index plate 814, appears in the entire image 200.

[0077] The indicator plate 814 is D-shaped. The indicator plate 814 is used to check history information related to the file creation process of the scanned files that have been created up to that point. As shown in FIG. 15A, the user places the indicator plate 814 in an empty space on the table 101 on which the document A 901 is placed. Then, as shown in FIG. 14B, the information display window 402 appears, notifying the user of the number of scanned files that have been created up to that point.

[0078] That is, the processing unit 163 counts the number of read files generated since the start of the work according to the "number of read files generated up to that point," which is history information previously set for the indicator board 813, and displays the information in the information display window 402. For example, as shown in the figure, it displays "15 files generated." When such a display is executed, the user can appropriately check the progress of the reading work. Note that the history information displayed in the information display window 402 is not limited to the number of read files generated. As will be described later, it may be configured so that the user can select the history information to be displayed.

[0079] 16A and 16B are diagrams for explaining the use of the index plate 815 and the processing set for the index plate 815. In particular, Fig. 16A is a diagram showing the state in which the index plate 815 is placed on the placement table 101 together with the index plate 800, and Fig. 16B is a diagram showing the state after the index plate region 300, which is the image of the index plate 800, and the index plate region 313, which is the image of the index plate 813, have appeared in the entire image 200. The processing set for the two index plates 800 is as described above.

[0080] The indicator board 815 is cross-shaped. The indicator board 815 is used to prohibit the creation of scanned files for all document areas. As shown in Fig. 16A, the user places two indicator boards 800 and one indicator board area 300 in an empty space on the table 101 where documents A 901, B 902, and C 903 are placed. Then, as shown in Fig. 16B, a warning display window 403 appears, notifying the user that the creation of scanned files is prohibited.

[0081] That is, the processing unit 163 superimposes a warning display window 403 indicating that the generation of a scanned file is prohibited on the entire image 200, and does not issue a file generation instruction even if it detects that two index boards 800 are simultaneously covered, as long as the index board area 315 is detected in the entire image 200. At this time, the frame indicators surrounding the document area A 201, the document area B 202, and the document area C 203 are not displayed. Note that the index board 815 may prohibit the execution of other setting processes in addition to or instead of the process of prohibiting the generation of a scanned file.

[0082] Next, the pre-setting of the setting process for each index board will be described. Fig. 17 is a diagram showing a pre-setting screen for the index board and the setting process. When using the document reading device 100, the user operates the operating members of the input unit 150 to call up the pre-setting screen shown in the figure and perform the setting.

[0083] As shown in the figure, the pre-setting screen is mainly composed of an index field 281 and a setting process field 282. The index field 281 shows at least the shape of each index. Depending on the index, the color, pattern, and text information are also displayed along with the shape. The setting process field 282 shows the setting process set for each index. In this embodiment, the setting process relates to the process of extracting the document area recognized by the recognition unit 161 from the entire image 200 and generating a read file, as explained using the above examples.

[0084] Initially, the setting process for each index is applied with a default setting. If the user wants to change the default setting, he or she can move the cursor 270 to the target setting in the setting process field 282 and click it, which will display a list of selectable setting processes, allowing the user to select the desired setting process. It is also possible to delete setting processes for index boards that the user does not want to use. It is also possible to register a new index board and set setting processes for that index board. For example, if a newly produced index board is photographed with the imaging unit 120 and the image is added to the index field 281, setting processes can be set for the new index board.

[0085] The user can display the next page's preset screen by clicking the page indicator 283. Figure 18 shows the preset screen for the second page of the indicator board and setting process. Like the first page, the second page is mainly composed of an indicator column 281 and a setting process column 282, but depending on the indicator board, the setting process column 282 may include a user specification column 284.

[0086] For example, for star-shaped index board 812, a specified character string field for specifying a character string to be detected is provided as user specification field 284, and the user types the text of the character string to be detected into the specified character string field. Also, for multi-projection index board 813, a specified logo registration field for specifying an image file of a logo to be detected is provided as user specification field 284, and the user calls up the image file of the logo to be specified from the image files stored in storage unit 170, for example.

[0087] For index plates that add additional information to a read file, the user can select what additional information to add in a user specification field 284. When the user places the cursor 270 on the user specification field 284 and clicks, a selection box 285 appears, and the user can select one of the additional information items provided in the selection box 285.

[0088] Next, the procedure of the document reading process executed by the control unit 160 will be described. Fig. 19 is a flow diagram showing the procedure of the control unit 160. The flow starts when the user places the document to be read on the table 101. It is assumed that the setting process for each index has already been completed.

[0089] In step S101, control unit 160 transmits an imaging request signal to imaging unit 120 to cause imaging processing to be performed, receives the imaging signal output from imaging unit 120, and causes image processing unit 180 to perform image processing to acquire an entire image. In step S102, recognition unit 161 recognizes the document areas of each of the multiple documents placed on table 101 from the acquired entire image. In step S103, control unit 160 generates a display signal in which a frame indicator or the like is superimposed on each document area recognized by recognition unit 161, and causes display unit 140 to convert the signal into a visible monitor image and display it.

[0090] In step S104, the processing unit 163 checks whether the recognition unit 161 has recognized the indicator board from the entire image 200. If the recognition unit 161 has recognized the indicator board, the process proceeds to step S105, and if the recognition unit 161 has not recognized the indicator board, the process skips to step S108.

[0091] If the process proceeds to step S105, the processing unit 163 confirms whether the index boards recognized by the recognition unit 161 are two index boards 800 and whether the processing unit 163 has recognized a user's hand gesture that covers both of them, that is, whether a file creation instruction has been received from the user. If it is confirmed that a file creation instruction has been received, the process proceeds to step S106, where the processing unit 163 extracts document areas from the entire image to generate multiple read files, performs an association process to associate these read files, and then proceeds to step S108. If it is confirmed in step S105 that the instruction is not a file creation instruction, the process proceeds to step S106, where the processing unit 163 executes a setting process that has been set for the index boards recognized by the recognition unit 161, and then proceeds to step S108.

[0092] In step S108, the control unit 160 checks whether an end instruction has been received from the user. If an end instruction has not been received, the control unit 160 returns to step S101 and repeats the series of processes. If an end instruction has been received, the series of processes is terminated.

[0093] Next, the usage modes of several index boards and the processing set for each index board will be described. Figure 20 is a diagram for explaining the usage modes of index board 816 and index board 817 and the processing set for each index board. In particular, Figure 20(A) is a diagram showing the state in which index board 816 and index board 817 are placed on table 101, and Figure 20(B) is a diagram showing the state after index board area 316, which is the image of index board 816, and index board area 317, which is the image of index board 817, have appeared in overall image 200.

[0094] The indicator plate 816 is shaped like an arrow and has the word "image" written on its surface. The indicator plate 816 is used to identify an image area present in the direction of the arrow within the document area and to perform image processing appropriate for the image on that image area when generating a scanned file. As shown in FIG. 20A , document C903 includes, for example, a photographic image. The user places the indicator plate 816 so that the arrow direction points toward the photographic image within document C903 placed on the table 101. Then, as shown in FIG. 20B , an image area 203a is recognized from the document area C203, and an image frame indicator 242 appears surrounding it. That is, the processing unit 163 searches for an image area 203a within the document area C203 that satisfies the conditions for extracting the image area from the direction indicated by the indicator plate area 316, generates an image frame indicator 242, and superimposes it on the entire image 200.

[0095] A condition for extracting an image region is, for example, that when a luminance histogram is created for each small area set in a matrix for the document region, the image region corresponds to a frequency distribution that indicates a certain characteristic. When a luminance histogram is created for each small area set in a matrix for the document region C203, the processing unit 163 recognizes the image region 203a as a region that includes small areas that correspond to the frequency distribution that indicates a certain characteristic of the image region. For example, while the character region of a typical document has a frequency biased toward luminance corresponding to white and black, an image region, typically a photograph, can be defined as having one or more gradual distributions in intermediate luminance.

[0096] The indicator board 817 is shaped like an arrow and has the word "diagram" written on its surface. The indicator board 817 is used to identify a diagram area present in the direction of the arrow within the document area and to perform image processing appropriate for the diagram on the diagram area when generating a scanned file. As shown in FIG. 20A, document A901 includes, for example, a diagram. The user places the indicator board 817 so that the arrow direction points toward the diagram within document A901 placed on the table 101. Then, as shown in FIG. 20B, the diagram area 201a is recognized from the document area A201, and a diagram frame indicator 241 appears surrounding it. That is, the processing unit 163 searches for the diagram area 201a within the document area A201 that satisfies the conditions for extracting the diagram area from the direction indicated by the indicator board area 317, generates the diagram frame indicator 241, and superimposes it on the overall image 200.

[0097] The condition for extracting a graphic area is, for example, that when an RGB histogram is created for each small area set in a matrix for the document area, the graphic area corresponds to a frequency distribution that indicates a characteristic of the graphic area. When an RGB histogram is created for each small area set in a matrix for the document area A 201, the processing unit 163 recognizes the graphic area 201a as an area that includes a small area that corresponds to the frequency distribution that indicates a characteristic of the graphic area. For example, while the character area of ​​a typical document has a frequency bias toward RGB values ​​corresponding to white and black, the graphic area can be defined as having a discrete distribution of RGB values ​​corresponding to primary colors.

[0098] When a file generation instruction is received from the user, the processing unit 163 cooperates with the image processing unit 180 to generate a read file by cutting out each area corresponding to the frame indicator from the entire image 200. At this time, the processing unit 163 generates the read file by applying processing differently to the specified image area and graphic area of ​​the document area than to other areas.

[0099] Specifically, when generating a scanned file corresponding to document area A201, processing unit 163 applies image processing appropriate for the graphic area 201a included in document area A201. For example, the saturation and contrast are increased compared to other areas. By generating a scanned file in this manner, the images and graphics included in the document can be made to look better. Furthermore, when generating a scanned file corresponding to document area C203, processing unit 163 applies image processing appropriate for the image to image area 203a included in document area C203. For example, a tone curve suitable for photographs is applied to increase the brightness and contrast compared to other areas. Image processing that makes the image more visible may be applied depending on the state of image area 203a actually scanned.

[0100] In the above example, the index boards 816 and 817 are placed on the table 101 at the same time. However, they may be used independently or together with other index boards. In the above example, the specific areas to be extracted are described as image areas and graphic areas. However, the specific areas extracted by the processing unit 163 according to the settings of the index boards are not limited to image areas and graphic areas. For example, the processing unit 163 may be configured to extract correction areas overwritten by the document creator so that the user can determine whether or not to generate a scanned file. Figure 21 is a diagram illustrating the use of the index board 818 and the processing set for the index board 818, particularly the processing by the processing unit 163 to extract correction areas from the target document area. Figure 21 (A) illustrates the user placing the index board 818 on the table 101, and Figure 21 (B) illustrates the state after the index board area 318, which is the image of the index board 818, appears in the overall image 200.

[0101] The indicator plate 818 is shaped like an eraser. The indicator plate 818 is used to identify correction areas within a document area and allow the user to recognize those areas. As shown in FIG. 21A , document A 901 includes a handwritten strikethrough line 901b that crosses a sentence, and document B 902 includes a correction fluid area 902a where correction fluid has been applied. The strikethrough line 901b and the correction fluid area 902b are both correction areas overwritten by the document creator. The user places the indicator plate 818 in an empty space on the table 101 on which documents A 901, B 902, and C 903, which are the objects for which the user wishes to check for the presence or absence of correction areas, are placed. 21B, a strikethrough area 201b is recognized from document area A201, and a strikethrough box indicator 243 appears surrounding it, while a correction fluid area 202a is recognized from document area B202, and a correction fluid box indicator 244 appears surrounding it. That is, the processing unit 163 extracts, from each document area, an area that meets the conditions for extracting a strikethrough area as a strikethrough area, and an area that meets the conditions for extracting a correction fluid area as a correction fluid area. Then, a strikethrough box indicator 243 is generated for the extracted strikethrough area, and a correction fluid box indicator 244 is generated for the extracted correction fluid area, and each is superimposed on the entire image 200.

[0102] The condition for extracting a strikethrough area can be defined, for example, as the presence of a straight line that passes continuously through the character area containing each character recognized by the OCR function. Because the whiteness of correction fluid is stronger than that of the base of the document area, the condition for extracting a correction fluid area can be defined, for example, as the frequency distribution being biased toward higher brightness than the frequency distribution of the area corresponding to the base of the document area when a brightness histogram is created for each small area set in a matrix for the document area. Note that a correction fluid area is not limited to an area where correction fluid has been applied, but can also be an area where correction tape has been applied. It is also preferable to recognize a correction fluid area even if characters have been added over the correction fluid or correction tape.

[0103] The user can set in advance whether the correction area to be extracted is the strikethrough area, the correction fluid area, or both. The example in Figure 21 shows a case where both the strikethrough area and the correction fluid area have been set as the correction area.

[0104] Furthermore, when a correction area is extracted, it is possible to set in advance whether to prohibit the generation of a read file only for that document area, to prohibit the generation of a read file including a document area that does not include a correction area, or to allow the generation of a read file regardless of whether a document area is extracted. Therefore, when a file generation instruction is received from the user in the state of Figure 21, the processing unit 163 generates a read file in accordance with the setting. Note that if the user instructs the generation of a read file even though the generation of a read file is prohibited, a notification window stating, for example, "A read file cannot be generated because a correction area exists" may be superimposed on the entire image 200.

[0105] 22A and 22B are diagrams for explaining the use of the index board 819 and the processing set for the index board 819. In particular, Fig. 22A is a diagram showing the state in which the index board 819 is placed on the table 101, and particularly shows the vicinity of the document A901, and Fig. 22B is a diagram showing the state after the index board area 319, which is the image of the index board 819, has appeared in the entire image 200, together with an enlarged view of a portion thereof.

[0106] The index plate 819 has a rectangular shape with one of its four corners chamfered. The index plate 819 is used to check, for each of the recognized document areas, whether a mark identical to the registered mark set is included in each of the recognized document areas. As shown in FIG. 22A, the user places the index plate 819 in an empty space on the placement table 101. Then, as shown in FIG. 22B, the system determines whether a mark area matching the registered mark set exists in the target document area. If such a mark area is found, a mark index appears surrounding the mark area. A notification window 543 is also displayed, indicating the number of registered marks identified for each registered mark. In the example of FIG. 14 described above, a logo was set in advance, and the processing unit 163 determined whether each document area contains a logo identical to the registered logo set. The example of FIG. 22 is similar to the example of FIG. 14. The user sets one or more marks to be checked for in the document, and the processing unit 163 determines, for each registered mark set, whether each document area contains a mark identical to the registered mark set.

[0107] When a document to be read includes, for example, a floor plan or equipment diagram of a property, a user may want to check whether a specific piece of equipment is included in the property before generating a scanned file. In such cases, the drawing symbol of the equipment to be checked is set in advance as a registered mark. Figure 11 shows an example in which document A901 to be read includes a floor plan of a property, and a toilet symbol 252a and a stove symbol 252b, which are the drawing symbols for a toilet and a stove, respectively, are set in advance as registered marks. Document B902 includes floor plans for the first to third floors of a property, and when a user wants to check whether a pre-set drawing symbol is included in such a document, the user places the indicator plate 819 on the table 101.

[0108] The processing unit 163 cooperates with the image processing unit 180 to determine whether the document area includes the toilet symbol 252a and the stove symbol 252b. The inclusion of these drawing symbols can be determined, for example, by using a pattern matching technique in which images of pre-registered image files of the toilet symbol 252a and the stove symbol 252b are applied to each document area to calculate the degree of match. In this case, the drawing symbols to be identified can be properly identified even if they exist in the document area in a different orientation or size than the image in the image file. Furthermore, determining whether a set registered mark is included in a document area is not limited to the pattern matching technique; other techniques, such as inputting the registered mark into a learning model and outputting the area corresponding to the registered mark, may also be used.

[0109] When the processing unit 163 finds the set drawing symbol within the document area, it superimposes a mark index on the entire image 200 to allow the user to visually identify the location of the drawing symbol, and displays the superimposed mark index on the display unit 140. Specifically, as shown in the figure, a mark index A 524a, which is a frame index surrounding the toilet symbol 252a, is superimposed on the toilet symbol 252a, and a mark index B 524b, which is a frame index surrounding the stove symbol 252b, is superimposed on the stove symbol 252b. The mark indexes are CG, and when multiple types of mark indexes are superimposed, such as the mark index A 524a and the mark index B 524b, it is preferable that they be distinguished from each other by color, for example.

[0110] The processing unit 163 also displays information regarding the number of identified drawing symbols on the display unit 140. Specifically, a notification window 543 is provided in an area that overlaps as little as possible with the document area, and the number of identified toilet symbols 252a and the number of identified stove symbols 252b are displayed in the notification window 543 along with the respective drawing symbols. At this time, it is advisable to also display in the notification window 543 an identification number for the document area in which each drawing symbol was found (in the illustrated example, "Document 1" corresponding to document area A201).

[0111] When generating a scanned file, the processing unit 163 may write information about the discrimination result in association with the scanned file to be generated, for example, in the tag area of ​​each file. For example, the tag area of ​​a scanned file generated from document area A201 in which toilet symbol 252a and stove symbol 252b were found may be written as "toilet symbol 252a: 3, stove symbol 252b: 2." Note that the processing unit 163 may determine, from the document area to be discriminated, an area of ​​a symbol legend in which set drawing symbols are listed and arranged, and exclude that area from the target area for counting the number of each drawing symbol.

[0112] 22 illustrates an example in which the floor plan symbols for a toilet and a stove are identified and displayed to the user from a property floor plan. However, the target floor plan symbols are not limited to these, and other floor plan symbols may also be set as registered marks. The number of registered marks that can be set may be one, three, or more. The floor plan may not only be included in document area A201, but may also be included in other document areas. In such cases, the processing unit 163 determines whether or not a registered mark has been set for each of the target document areas. The processing unit 163 may display the number of identified registered marks as the total number identified in all target document areas, or may display them separately for each document area.

[0113] Furthermore, the target registered marks are not limited to drawing symbols used on floor plans of properties, but can also be marks such as signs, emblems, signs, seal impressions, etc. Therefore, the objects for which the presence or absence of these marks can be determined are not limited to floor plans, but can also include, for example, mechanical drawings, circuit diagrams, maps, mathematical formulas, chemical formulas, etc.

[0114] Fig. 23 is a modified example of the example of Fig. 22, and is a diagram for explaining the usage of the index board 820 and the processing set for the index board 820. In particular, Fig. 23(A) is a diagram showing the state in which the user places the index board 820 toward a specific location on the document A901 placed on the table 101, and Fig. 23(B) is a diagram showing the state after the index board area 320, which is the image of the index board 820, appears in the entire image 200.

[0115] The indicator plate 820 is shaped like a combination of an arrow and a circle, with the letter "M" written in the circle. In the example of Fig. 22, the indicator plate 819 was designed to identify all of a plurality of pre-set registered marks from the document area, but the indicator plate 820 is designed to identify a specific mark that exists in the direction indicated by the arrow among a plurality of pre-set registered marks, and further to extract only marks identical to the specific mark from the document area.

[0116] For example, if a user wants to know the number of toilet symbols contained in a target document or the locations of all toilet symbols, the user places the indicator plate 820 with the arrow pointing toward one of the toilet symbols contained in a floor plan, as shown in FIG. 23A. The processing unit 163 then determines whether a mark area matching the toilet symbol 252a, one of the pre-set registered marks, exists within the document area in the direction indicated by the arrow. If the processing unit 163 finds such a mark area, it then determines whether any other document area contains a mark area matching the toilet symbol 252a. Then, as shown in FIG. 23B, the processing unit 163 displays a mark indicator 525 superimposed on each of the determined mark areas.

[0117] Furthermore, the processing unit 163 displays information regarding the number of identified drawing symbols on the display unit 140. Specifically, a notification window 543 is provided in an area that overlaps as little as possible with the document area, and the number of identified toilet symbols 252a is displayed together with the drawing symbols in the notification window 543. At this time, it is advisable to also display in the notification window 543 an identification number for the document area in which the toilet symbol 252a was found (in the illustrated example, "Document 1" corresponding to document area A201).

[0118] 24 is a diagram showing the third page of the preset setting screen for the indicator board and setting process, following on from FIG. 18. Like the first and second pages, the third page is mainly composed of an indicator column 281 and a setting process column 282, but depending on the indicator board, the setting process column 282 may include a user specification column 284. For example, for the eraser-shaped indicator board 818, the user specification column 284 may include a specification box for specifying the target of the correction area, or a selection box for selecting whether or not to allow the generation of a read file when the correction area is extracted.

[0119] Additionally, for the rectangular index plate 819 with one of the four corners chamfered, a settings button 286 is provided in the user specification field 284. When the user clicks the settings button 286, a settings window 291 opens. The settings window 291 mainly contains an upload button 292, a registered mark list 293, radio buttons 294, and a done button 295. When the user clicks the upload button 292, the user can select an image file (e.g., a GIF file) containing marks. The marks contained in the selected image file are displayed in the registered mark list 293. The radio buttons 294 are arranged for selecting an option menu. In the illustrated example, clicking the "Count Display" radio button 294 displays the number of detected registered marks in the notification window 543. Clicking the "Allow file generation when no registered marks are detected" radio button 294 allows the creation of a scanned file even if no registered marks are detected. Clicking the done button 295 confirms the settings and closes the settings window 291.

[0120] The above-described embodiment shows only a limited number of examples of indicator boards and the setting processes set for those indicator boards. There are also a wide variety of indicator boards and setting processes set for each indicator board. For example, even if the indicator boards have the same shape, various setting processes can be set for each indicator board by changing the color, pattern, or text information written on the indicator board. Furthermore, if the orientation of an indicator board can be defined by its shape, various setting processes can be set depending on the position and orientation of the indicator board.

[0121] Furthermore, a set process can be set for a combination of multiple indicator boards, as with indicator board 800. Various variations can also be applied to the detection of hand gestures relative to the indicator boards. For example, when it is detected that a hand passes over a certain indicator board twice in succession, the process set for that indicator board can be executed.

[0122] Furthermore, in the present embodiment described above, the case where multiple documents are read has been mainly described, but of course the document reading device 100 can also generate a reading file when a single document is placed on the placement table 101.

Claims

1. A mounting table for mounting a document and an index plate, an imaging unit that images the mounting table and outputs an overall image of the mounting table, a recognition unit that recognizes a document area of the document placed on the mounting table and an index plate area of the index plate from the overall image, a processing unit that executes setting processing related to reading of the document area preset in association with at least any one of the shape, position, and orientation of the index plate based on at least any one of the shape, position, and orientation of the index plate area recognized by the recognition unit A document reading device comprising:

2. The document reading device according to claim 1, wherein the processing unit executes the setting processing preset for the index plate when the recognition unit recognizes a hand gesture performed by a user on the index plate on the mounting table.

3. The document reading device according to claim 1, wherein the processing unit executes the setting processing preset for the plurality of index plates when the recognition unit recognizes a combination of hand gestures performed by a user on the plurality of index plates on the mounting table.

4. The document reading device according to claim 1, wherein the processing unit executes the setting processing preset for the index plate based on at least any one of the color, pattern, and character information of the index plate area recognized by the recognition unit.

5. The document reading device according to claim 1, wherein the setting processing includes file generation processing for cutting out the document area from the overall image and generating a reading file.

6. The document reading device according to claim 1, wherein the setting processing includes selection processing for selecting a specific document area from a plurality of document areas recognized by the recognition unit when a plurality of documents are placed on the mounting table.

7. The document reading device according to claim 1, wherein the setting processing includes extraction processing for extracting a description frame included in the document area.

8. The document reading device according to claim 1, wherein the setting processing includes change processing for changing the document area.

9. The document reading device according to claim 1, wherein the setting processing includes addition processing for adding to a reading file for generating preset information.

10. The document reading device according to claim 1, wherein the setting processing includes history display processing for displaying history information related to file generation processing of a reading file generated so far.

11. The document reading device according to claim 1, wherein the setting processing includes prohibition processing for prohibiting execution of other processing.

12. The document reading apparatus according to claim 1, wherein the setting process includes a determination process for determining whether it matches information set in advance.

13. The document reading apparatus according to any one of claims 1 to 12, further comprising a reception unit that receives a setting of the setting process for the index plate from a user.

14. A document reading program for a document reading apparatus, comprising a placement table on which a document and an index plate are placed, and an imaging unit that images the placement table and outputs an overall image of the placement table, a recognition step of recognizing a document area of the document placed on the placement table and an index plate area of the index plate from the overall image, a processing step of executing setting processing related to reading of the document area preset in association with at least any one of the shape, position, and orientation of the index plate area based on at least any one of the shape, position, and orientation of the index plate area recognized in the recognition step A document reading program for causing a computer to execute.

15. The document reading apparatus according to claim 1, wherein the setting process includes an extraction process of extracting a specific area that satisfies an extraction condition set in association with the index plate from the document area.

16. The document reading apparatus according to claim 15, wherein when the setting process executes a file generation process of cutting out the document area from the overall image to generate a reading file, different processing is performed on the specific area among the document areas compared to other areas.

17. The document reading apparatus according to claim 15, wherein when the setting process executes a file generation process of cutting out the document area from the overall image to generate a reading file, a reading file is not generated for the document area including the specific area.

18. The document reading apparatus according to claim 15, wherein the extraction condition is a condition for extracting an image area.

19. The document reading apparatus according to claim 15, wherein the extraction condition is a condition for extracting a correction area overwritten by a document creator.

20. The document reading apparatus according to claim 12, wherein the determination process determines a mark area that matches the mark as the information from the document area.

21. As the information, a plurality of types of the marks can be set, The document reading apparatus according to claim 20, wherein the processing unit displays a mark indicator on a display unit so that a user can distinguish and visually recognize each type of the mark.

22. The document reading apparatus according to claim 20, wherein the processing unit displays information regarding the number of the marks discriminated by the discrimination process on a display unit.

23. The document reading apparatus according to claim 20, wherein when the discrimination processing unit discriminates one mark area indicated by the index plate area from the document area, the discrimination processing unit further searches for the mark areas of the same type from the document area.