Image processing device, image processing method, and image processing program

JP7877878B2Active Publication Date: 2026-06-23NEC CORP

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
NEC CORP
Filing Date
2022-06-28
Publication Date
2026-06-23

AI Technical Summary

Benefits of technology

【0009】 本発明の一態様によれば、複数の記入用紙を予め用意することなく、記入用紙の記入内容をより精度よく読み取ることができる。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007877878000001
    Figure 0007877878000001
  • Figure 0007877878000002
    Figure 0007877878000002
  • Figure 0007877878000003
    Figure 0007877878000003
Patent Text Reader

Abstract

To make it possible to more accurately read entry content of an entry form without preparing a plurality of entry forms in advance.SOLUTION: An image processing device (1) includes: a reading unit (11) for referring to a reference image representing a blank entry field to read entry content from an input image representing an entry form including an entry field or a partial image representing the entry field; a generation unit (12) for referring to a partial image group to which the partial image has been added to generate a new reference image representing a blank entry field; and an update unit (13) for replacing the reference image by a new reference image.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Claims

1. A reading means that reads the contents of an entry form, or a partial image representing the entry field, by referring to a reference image representing an unfilled entry field, Extraction means for extracting a partial image representing an unfilled input field from the input image or the partial image, A generation means that generates a new reference image representing an empty input field by referring to a group of partial images to which extracted partial images representing empty input fields have been added, An update means for replacing the aforementioned reference image with the aforementioned new reference image, A correction means for correcting the entered content in response to user operations, An extraction means for extracting individual feature quantities, which are feature quantities related to the entered content, from the corrected entered content, A calculation means that refers to the group of individual features to which the aforementioned individual features have been added, and calculates common features which are features of the content to be written on an entry form that is the same format as the entry form, The system includes a cropping means that crops a portion of an image from an input image by referring to the aforementioned common features, Image processing device.

2. A setting means that performs a setting process to set format information for the aforementioned input field in response to user operation, The system further includes cropping means that perform cropping operations to cut out the partial image from the input image by referring to the format information. The image processing apparatus according to claim 1.

3. Depending on whether the first input image in a series of input images representing an input form of the same format as the aforementioned input form represents an input image representing a completed input form or an input image representing an uncompleted input form, the cutting means, the generating means, and the updating means each decide whether or not to perform the above processing on the second and subsequent input images. The image processing apparatus according to claim 2.

4. Depending on the similarity between the feature quantities relating to the entered content read by the reading means and the common feature quantities, the correction means, the extraction means, the calculation means, and the clipping means each decide whether or not to perform the above processing. The image processing apparatus according to claim 3.

5. At least one processor, The process involves referring to a reference image representing an unfilled entry field, reading the content from an input image representing an entry form including the entry field, or from a partial image representing the entry field, Extracting a partial image representing an unfilled input field from the input image or the partial image, The extracted partial images representing the blank fields are referenced to the added partial image group to generate a new reference image representing the blank fields, and Replacing the aforementioned reference image with the aforementioned new reference image, The above-mentioned content will be modified in response to user actions, From the corrected entries, extract individual features that are characteristic quantities related to the entries, The individual features are added to the group of individual features, and a common feature is calculated, which is the feature of the content written on a form with the same format as the aforementioned form. This includes, by referring to the aforementioned common features, cutting out the partial image from the input image, Image processing methods.

6. Computers, A reading means that reads the contents of an entry form, or a partial image representing the entry field, by referring to a reference image representing an unfilled entry field, Extraction means for extracting a partial image representing an unfilled input field from the input image or the partial image, A generation means that generates a new reference image representing an empty input field by referring to a group of partial images to which extracted partial images representing empty input fields have been added, An update means for replacing the aforementioned reference image with the aforementioned new reference image, A correction means for correcting the entered content in response to user operations, An extraction means for extracting individual feature quantities, which are feature quantities related to the entered content, from the corrected entered content, A calculation means that refers to the group of individual features to which the aforementioned individual features have been added, and calculates common features which are features of the content to be written on an entry form that is the same format as the entry form, The aforementioned common features are referenced and the resulting cropping means is used to cut out the partial image from the input image. Image processing program.