Image processing apparatus and image processing method
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- KYOCERA DOCUMENT SOLUTIONS INC
- Filing Date
- 2026-01-13
- Publication Date
- 2026-07-30
Smart Images
Figure JP2026000635_30072026_PF_FP_ABST
Abstract
Description
Image Processing Apparatus and Image Processing Method
[0001] The present invention relates to an image processing apparatus and an image processing method.
[0002] A certain image processing apparatus performs character recognition processing different from printed character strings on handwritten character strings described in a document as item values for specific items (see, for example, Patent Document 1).
[0003] Japanese Unexamined Patent Application Publication No. 2022-116983
[0004] However, it is difficult for the above-described image processing apparatus to confirm that personal identification information such as a name written as a handwritten character string is legitimate (that is, the person who wrote the handwritten characters matches the person of the personal identification information).
[0005] The present invention has been made in view of the above problems, and an object thereof is to obtain an image processing apparatus and an image processing method capable of confirming that personal identification information written as a handwritten character string is legitimate.
[0006] The image processing apparatus according to the present invention includes a handwritten character string detection unit that detects a handwritten character string from a predetermined area in a document image, and (a) acquires verification personal identification information from a specific recording medium, and (b) verifies whether the handwritten character string detected from the area where the personal identification information should be entered matches the verification personal identification information, and (c) when the detected handwritten character string matches the verification personal identification information, determines that the detected handwritten character string is legitimate, and when the detected handwritten character string does not match the verification personal identification information, determines that the detected handwritten character string is not legitimate, and a handwritten character string verification unit.
[0007] The image processing method according to the present invention comprises the steps of: (a) obtaining verification personal identification information from a specific recording medium; (b) verifying whether the handwritten string detected from the area where the personal identification information should be written matches the verification personal identification information; and (c) determining that the detected handwritten string is valid if it matches the verification personal identification information, and determining that the detected handwritten string is invalid if it does not match the verification personal identification information.
[0008] The image processing program according to the present invention causes a computer to function as the handwritten character detection unit and the handwritten character verification unit described above.
[0009] Furthermore, the image processing apparatus according to the present invention includes a handwritten character detection unit that detects a handwritten character from a predetermined area in a document image, and a handwritten character verification unit that, when a correction object is detected together with the handwritten character, (a) if a correction stamp object is detected together with the correction object, determines that the detected handwritten character is valid, and (b) if a correction stamp object is not detected together with the correction object, determines that the detected handwritten character is invalid.
[0010] Furthermore, the image processing apparatus according to the present invention includes a handwritten string detection unit that detects a handwritten string from a predetermined area in a document image, and a handwritten string verification unit that determines that the detected handwritten string is not valid if an overlaid member object is detected together with the handwritten string.
[0011] According to the present invention, an image processing apparatus and an image processing method are obtained that can verify that personal identification information written as a handwritten string is legitimate.
[0012] The above or other objects, features, and advantages of the present invention will become even more apparent from the following detailed description in conjunction with the accompanying drawings.
[0013] Figure 1 is a block diagram showing the configuration of an image processing apparatus according to an embodiment of the present invention. Figure 2 is a diagram showing an example of a form area and handwritten text in a document image. Figure 3 is a flowchart illustrating the operation of the image processing apparatus shown in Figure 1. Figure 4 is a diagram showing an example of a correction object and a correction stamp object in Embodiment 2. Figure 5 is a diagram showing an example of a handwritten text and an overlay member object in Embodiment 4.
[0014] Embodiments of the present invention will be described below with reference to the figures.
[0015] Embodiment 1.
[0016] Figure 1 is a block diagram showing the configuration of an image processing apparatus according to an embodiment of the present invention. The image processing apparatus shown in Figure 1 is an information processing apparatus such as a personal computer or a server, or an electronic device such as a digital camera or an image forming apparatus (scanner, multifunction printer, etc.), and comprises a processing unit 1, a storage device 2, a communication device 3, a display device 4, an input device 5, an internal device 6, and the like.
[0017] The arithmetic processing unit 1 includes a computer and operates as various processing units by executing image processing programs on that computer. Specifically, the computer includes a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), etc., and operates as a predetermined processing unit by loading programs stored in the ROM or storage device 2 into the RAM and executing them with the CPU. The arithmetic processing unit 1 may also include an ASIC (Application Specific Integrated Circuit) that functions as a specific processing unit.
[0018] The storage device 2 is a non-volatile storage device such as flash memory, and stores the image processing program 2a and data (such as form data 2b) necessary for the processing described later. The image processing program 2a is stored, for example, on a non-temporary, computer-readable recording medium and installed from that recording medium to the storage device 2.
[0019] Form data 2b is data that indicates the position and size of a form area within a specific target image where a user has entered handwritten text for a specific item.
[0020] Communication device 3 is a device that communicates data with external devices (such as an IC card reader or external server, as described later), and is such as a network interface or peripheral device interface. Display device 4 is a device that displays various information to the user, and is such as a display panel such as an LCD display. Input device 5 is a device that detects user operations, and is such as a keyboard or touch panel.
[0021] The internal device 6 is a device that performs a predetermined function of the image processing device. For example, if the image processing device is an image forming apparatus, the internal device 6 may be an image reading device that optically reads an image from a document, or a printing device that prints an image onto printing paper.
[0022] In this configuration, the arithmetic processing unit 1 operates based on the image processing program 2a as the aforementioned processing units: the target image acquisition unit 11, the form area identification unit 12, the handwritten character detection unit 13, and the handwritten character verification unit 14.
[0023] The target image acquisition unit 11 acquires document images as target images (raster image data) from the storage device 2, communication device 3, internal device 6, etc., and stores them in RAM or the like.
[0024] For example, this document image may be a read image of a document (such as an application form, roster, or contract) in which the handwritten text to be extracted is written in the form area defined by form data 2b, or it may be an image obtained by superimposing handwritten text entered through user operation (entry operation) on a touch panel or the like onto a document image of a document (such as an application form, roster, or contract) displayed on a tablet or the like.
[0025] The form area identification unit 12 identifies the form area (specifically, its position and size) in the document image based on the form data 2b. Here, the form area is a rectangular area in which the string to be extracted is written.
[0026] The handwritten character detection unit 13 detects handwritten characters written in the form area of the document image.
[0027] Figure 2 shows an example of a form area and handwritten text in a document image.
[0028] Specifically, the handwritten string detection unit 13 detects handwritten string areas in a form area, for example, as shown in Figure 2, performs character recognition processing on the handwritten string areas, and extracts the handwritten strings. For example, the handwritten string detection unit 13 detects handwritten string areas using an existing object detection method that uses a machine learning-trained learner, and extracts the handwritten strings by performing existing character recognition processing that uses a machine learning-trained learner.
[0029] The handwritten string verification unit 14 (a) obtains personal identification information for verification from a specific recording medium, (b) verifies whether the handwritten string detected from the area where personal identification information should be entered matches the personal identification information for verification, and (c) if the detected handwritten string matches the personal identification information for verification, it determines that the detected handwritten string is valid, and if the detected handwritten string does not match the personal identification information for verification, it determines that the detected handwritten string is invalid.
[0030] If the detected handwritten string is determined to be valid, the handwritten string verification unit 14 outputs the text data of the detected handwritten string to a subsequent process or another data processing system. On the other hand, if the detected handwritten string is determined to be invalid, the handwritten string verification unit 14 either (a) discards the detected handwritten string or (b) outputs the text data of the detected handwritten string along with the verification result to a subsequent process or another data processing system.
[0031] The specific recording medium mentioned above is, for example, an IC card (such as a My Number Card or driver's license) that contains personal identification information for verification, and the handwritten character verification unit 14 reads the personal identification information for verification from the IC card using an IC card reader.
[0032] Here, personal identification information and verification personal identification information are information that includes at least a name. Personal identification information and verification personal identification information may also include current address, individual number (My Number), etc.
[0033] Furthermore, in this embodiment, the handwritten string detection unit 13 may detect superimposed member objects along with the handwritten string area in the form area, and the handwritten string verification unit 14 may determine that the detected handwritten string is invalid if it detects a superimposed member object (an image of a correction member used to superimpose on handwritten strings, such as a sticky note or correction tape) along with the handwritten string.
[0034] For example, the handwritten character detection unit 13 detects superimposed member objects using an existing object detection method that employs a machine learning-trained learner. Superimposed member objects are detected by a machine learning learner that has been trained using images containing superimposed member objects of various forms as training data.
[0035] Furthermore, in this embodiment, the handwritten string verification unit 14 may obtain handwriting information as verification personal identification information associated with the name indicated by the detected handwritten string from an external server or the like using the communication device 3, and if the handwriting information of the handwritten string does not match the obtained handwriting information, it may determine that the detected handwritten string is not valid. Note that the handwriting information is data that shows feature quantities for predetermined items such as the shape of the characters and the thickness of the characters, and is pre-registered on an external server or the like in association with the name of each individual.
[0036] Next, the operation of the image processing device shown in Figure 1 will be explained. Figure 3 is a flowchart illustrating the operation of the image processing device shown in Figure 1.
[0037] When the target image acquisition unit 11 acquires a target image (document image) (step S1), the form area identification unit 12 identifies the position and size of the form area in the document image based on the form data 2b corresponding to the document image (step S2), the handwritten string detection unit 13 detects the handwritten string area in the form area (step S3), performs character recognition processing on the handwritten string area, and extracts the handwritten string (text data) (step S4).
[0038] The handwritten string verification unit 14 uses the communication device 3 to obtain verification personal identification information (text data such as a name) from a specific recording medium (for example, the IC card of the person who wrote the handwritten string) (step S5), and determines whether the handwritten string matches the verification personal identification information (step S6).
[0039] If the handwritten string described above is determined to match the personal identification information for verification, the handwritten string verification unit 14 determines that the handwritten string is valid (step S7) and outputs data containing the handwritten string (text data) corresponding to the document image (step S8).
[0040] On the other hand, if the handwritten string does not match the personal identification information used for verification, the handwritten string verification unit 14 determines that the handwritten string is not valid (step S9) and performs error processing (discarding the handwritten string, notifying the user, etc.) (step S10). Alternatively, instead of error processing, the verification result may be attached to the data output.
[0041] As described above, according to the first embodiment, the handwritten character string detection unit 13 detects a handwritten character string from a predetermined area in the document image. The handwritten character string verification unit 14 (a) acquires verification personal identification information from a specific recording medium, (b) verifies whether the handwritten character string detected from the area where the personal identification information should be entered matches the verification personal identification information, and (c) if the detected handwritten character string matches the verification personal identification information, determines that the detected handwritten character string is valid, and if the detected handwritten character string does not match the verification personal identification information, determines that the detected handwritten character string is not valid.
[0042] Thereby, it can be confirmed that the personal identification information entered as a handwritten character string is legitimate.
[0043] Second Embodiment.
[0044] FIG. 4 is a diagram showing an example of a correction object and a correction mark object in the second embodiment. In the second embodiment, when a correction object is detected together with a handwritten character string in a form area, for example, as shown in FIG. 4, the handwritten character string verification unit 14 (a) when a correction mark object is detected together with the correction object, determines that the detected handwritten character string is valid if the detected handwritten character string matches the verification personal identification information, and (b) when a correction mark object is not detected together with the correction object, determines that the detected handwritten character string is not valid even if the detected handwritten character string matches the verification personal identification information.
[0045] For example, the handwritten character string detection unit 13 detects a handwritten character string area and a correction object by using an existing object detection method with a machine-learned learner, and performs an existing character recognition process using the machine-learned learner to extract a handwritten character string from the handwritten character string area.
[0046] The correction object is an object obtained by superimposing a strikethrough or the like on a handwritten character, and is detected by a learner machine-learned using an image including correction objects in various forms as training data.
[0047] Furthermore, in the second embodiment, the handwritten character verification unit 14 acquires, as verification personal identification information, correction mark information (for example, a stamped image associated with the name detected as a handwritten character string) from an external server or the like using the communication device 3. When the detected handwritten character string matches the verification personal identification information and the correction mark object matches the correction mark information, it may be determined that the detected handwritten character string is valid. When the correction mark object does not match the correction mark information, even if the detected handwritten character string matches the verification personal identification information, it may be determined that the detected handwritten character string is not valid.
[0048] Furthermore, in the second embodiment, the handwritten character verification unit 14 may determine that the detected handwritten character string is not valid when the color of the handwritten character string and the color of the correction object (at least a part of the correction object) are different from each other.
[0049] Furthermore, in the second embodiment, the handwritten character verification unit 14 may determine that the detected handwritten character string is not valid when the thickness of the characters in the handwritten character string and the thickness of the characters in the correction object are different from each other.
[0050] That is, in these cases, since the writing instrument that filled in the original character string canceled by the correction object is different from the writing instrument that filled in the detected handwritten character string, it is presumed that the person who filled in the original character string canceled by the correction object is different from the person who filled in the detected handwritten character string, and it is determined that the detected handwritten character string is not valid.
[0051] Note that other configurations and operations of the image processing apparatus according to the second embodiment are the same as those of the first embodiment, and thus the description thereof is omitted.
[0052] Embodiment 3.
[0053] In Embodiment 3, when a correction object is detected along with the handwritten string, the handwritten string verification unit 14 determines that (a) the detected handwritten string is valid if a correction stamp object is detected along with the correction object, and (b) the detected handwritten string is invalid if no correction stamp object is detected along with the correction object. In Embodiment 3, the handwritten string is not limited to a string of personal identification information (such as a name).
[0054] Furthermore, the other configurations and operations of the image processing device according to Embodiment 3 are the same as those of any of the other embodiments, so their description will be omitted.
[0055] Embodiment 4.
[0056] Figure 5 shows an example of a handwritten string and an overlaid member object in Embodiment 4. In Embodiment 4, the handwritten string verification unit 14 determines that the detected handwritten string is invalid if an overlaid member object is detected along with the handwritten string, as shown in Figure 5, for example. In Embodiment 4, the handwritten string is not limited to a string of personal identification information (such as a name). For example, in the case of Figure 5, a handwritten string of an amount of money is extracted.
[0057] Furthermore, the other configurations and operations of the image processing device according to Embodiment 4 are the same as those of any of the other embodiments, so their description will be omitted.
[0058] Furthermore, various changes and modifications to the embodiments described above will be obvious to those skilled in the art. Such changes and modifications may be made without deviating from the spirit and scope of the subject matter and without diminishing the intended advantages. In other words, such changes and modifications are intended to be included in the claims.
[0059] For example, in the above embodiment, the image processing device is a device used directly by the user, but it may also be a device such as a server (e.g., a cloud server) used indirectly by the user via a network. In that case, the user interface of the client device is used instead of the display device 4 and the input device 5.
[0060] The present invention is applicable, for example, to an image processing apparatus.
Claims
1. An image processing apparatus comprising: a handwritten character detection unit that detects a handwritten character from a predetermined area in a document image; and a handwritten character verification unit that (a) acquires personal identification information for verification from a specific recording medium; (b) verifies whether the handwritten character detected from the area where the personal identification information should be written matches the personal identification information for verification; and (c) determines that the detected handwritten character is valid if it matches the personal identification information for verification, and determines that the detected handwritten character is invalid if it does not match the personal identification information for verification.
2. The image processing apparatus according to claim 1, characterized in that the specific recording medium is an IC card containing the personal identification information for verification.
3. The image processing apparatus according to claim 1, wherein, when a correction object is detected together with the handwritten string, (a) when a correction stamp object is detected together with the correction object, the detected handwritten string is determined to be valid if the detected handwritten string matches the verification personal identification information, and (b) when no correction stamp object is detected together with the correction object, the detected handwritten string is determined to be invalid even if the detected handwritten string matches the verification personal identification information.
4. The image processing apparatus according to claim 3, wherein the handwritten string verification unit acquires correction seal information as the personal identification information for verification, and determines that the detected handwritten string is not valid if the correction seal object does not match the correction seal information.
5. The image processing apparatus according to claim 3, characterized in that the handwritten string verification unit determines that the detected handwritten string is not valid if the color of the handwritten string and the color of the correction object are different from each other, or if the thickness of the characters in the handwritten string and the thickness of the characters in the correction object are different from each other.
6. The image processing apparatus according to claim 1, characterized in that, when the handwritten string verification unit detects an overlaid member object together with the handwritten string, it determines that the detected handwritten string is not valid.
7. The image processing apparatus according to claim 1, characterized in that the handwritten string verification unit acquires handwriting information as personal identification information for verification, and determines that the detected handwritten string is not valid if the handwriting information of the handwritten string does not match the acquired handwriting information.
8. An image processing method comprising: (a) detecting a handwritten string from a predetermined area in a document image; (b) obtaining personal identification information for verification from a specific recording medium; (c) verifying whether the handwritten string detected from the area where the personal identification information should be written matches the personal identification information for verification; and (d) determining that the detected handwritten string is valid if it matches the personal identification information for verification, and determining that the detected handwritten string is invalid if it does not match the personal identification information for verification.
9. An image processing apparatus comprising: a handwritten character detection unit that detects a handwritten character from a predetermined area in a document image; and a handwritten character verification unit that, when a correction object is detected together with the handwritten character, (a) determines that the detected handwritten character is valid if a correction stamp object is detected together with the correction object; and (b) determines that the detected handwritten character is invalid if no correction stamp object is detected together with the correction object.
10. An image processing apparatus comprising: a handwritten character detection unit that detects a handwritten character from a predetermined area in a document image; and a handwritten character verification unit that determines that the detected handwritten character is not valid when an overlaid member object is detected together with the handwritten character.