Image processing system, image processing method, and program

Patent metadata

Field	Value
Publication number	US-12597280-B2
Application number	US-202117800891-A
Country	US
Kind code	B2
Filing date	Jul 28, 2021
Priority date	Jul 28, 2021
Publication date	Apr 7, 2026
Grant date	Apr 7, 2026

Abstract

At least one processor of an image processing system acquires a target object image including a target object. The at least one processor executes matching for the target object image based on a template image including a feature about the target object. The at least one processor analyzes a plurality of scores acquired through the matching. The at least one processor determines, based on a result of analysis of the plurality of scores, whether the target object in the target object image is blurred.

First claim

The invention claimed is:

1 . An image processing system, comprising at least one processor configured to:
acquire a target object image including a target object; execute matching for a first region of pixels of the target object image based on a second region of pixels of a template image including a feature about the target object;
wherein the template image is clear;
analyze a plurality of scores indicating a degree of similarity between each pixel of the first region and the second region acquired through the matching; determine whether the target object in the target object image is blurred, based on a result of the analysis of the plurality of scores; not execute a further processing of the target image when the target object is determined to be blurred; wherein four regions in each of the template image and the target image undergo execution matching; wherein the at least one processor is configured to analyze the plurality of scores acquired for each of a plurality of template images by calculating one index based on the plurality of scores acquired for each template image; and wherein the at least one processor is configured to determine whether the target object in the target object image is blurred by comparing the index to a blur determination threshold.

2 . The image processing system according to claim 1 ,
wherein the target object includes one or more standard parts, wherein the feature included in the template image is the standard part, and wherein the plurality of scores each indicate a degree of similarity to the standard part included in the template image.

3 . The image processing system according to claim 2 ,
wherein the target image includes a plurality of the standard parts, wherein template images are prepared for each of the plurality of the standard parts, and wherein the at least one processor is configured to execute the matching for each template image.

4 . The image processing system according to claim 2 ,
wherein the standard part is placed near a corner of the target object, and wherein the at least one processor is configured to execute the matching for a region near the corner of the target object in the target object image.

5 . The image processing system according to claim 1 , wherein the at least one processor is configured to analyze the plurality of scores based on a highest score out of the plurality of scores and on a mean score calculated from the plurality of scores.

6 . The image processing system according to claim 5 , wherein the at least one processor is configured to analyze the plurality of scores based on a value that is a quotient of division of the highest score by the mean score.

7 . The image processing system according to claim 1 , wherein the at least one processor is configured to determine whether a highest score out of the plurality of scores is less than a threshold value, and
wherein, based on a result of determination relating to the highest score and a result of determination relating to the blurring, predetermined processing is executed.

8 . The image processing system according to claim 1 , wherein the at least one processor is configured to determine whether a distance between a position at which a highest score out of the plurality of scores is acquired and a position at which the feature is supposed to be located is less than a threshold value, and
wherein, based on a result of determination relating to the distance and a result of determination relating to the blurring, predetermined processing is executed.

9 . The image processing system according to claim 1 , wherein the at least one processor is configured to process at least a part of the target object image so that an outline of the target object in the target object image approximates a predetermined outline, and
wherein the at least one processor is configured to execute the matching for the target object image processed at least partially.

10 . The image processing system according to claim 1 , wherein the at least one processor is configured to determine whether the target object is a predetermined target object based on the target object image determined to be unblurred.

11 . The image processing system according to claim 10 , further comprising a user terminal and a server,
wherein the user terminal is configured to:
acquire the target object image;
execute the matching;
analyze the plurality of scores;
determine whether the target object is blurred; and
transmit, to the server, the target object image determined to be unblurred, and
wherein the server is configured to:
receive, from the user terminal, the target object image determined to be unblurred;
determine whether the target object is a predetermined target object; and
demand, when it is determined that the target object is not the predetermined target object, that the user terminal acquire the target object image again.

12 . The image processing system according to claim 1 ,
wherein the target object is an identity verification document of a user, wherein the target object image is an image generated by a camera by photographing the identity verification document in succession, wherein the at least one processor is configured to:
acquire the images generated in succession as the target object image,
execute the matching for each of the images generated in succession as the target object image,
analyze the plurality of scores for each of the images generated in succession as the target object image, and
determine, for each of the images generated in succession as the target object image, whether the identity verification document in the target object image is blurred.

13 . The image processing system according to claim 1 , wherein a number of scores of the plurality of scores is based on a first number of pixels in the first region and a second number of pixels in the second region.

14 . The image processing system according to claim 13 , wherein the number of pixels in the first region and the number of pixels in the second region are different.

15 . The image processing system according to claim 1 , wherein each of the regions of the four regions in the template image is located in one of four corners of the template image.

16 . An image processing method, comprising:
acquiring a target object image including a target object; executing matching for a first region of pixels the target object image based on a second region of pixels of a template image including a feature about the target object; wherein the template image is clear; analyzing a plurality of scores indicating a degree of similarity between each pixel of the first region and the second region acquired through the matching; determining whether the target object in the target object image is blurred, based on a result of the analysis of the plurality of scores;  not execute a further processing of the target image when the target object is determined to be blurred;  wherein four regions in each of the template image and the target image undergo execution matching;  analyzing the plurality of scores acquired for each of a plurality of template images by calculating one index based on the plurality of scores acquired for each template image; and  determining whether the target object in the target object image is blurred by comparing the index to a blur determination threshold.

17 . A non-transitory computer-readable information storage medium for storing a program for causing a computer to:
acquire a target object image

Assignees

• Rakuten Group Inc

Inventors

• CHAE Yeongnam

Classifications

• G06T2207/30176

  Document · CPC title

• G06T2207/30168

  Image quality inspection · CPC title

• G06T7/0002

  Inspection of images, e.g. flaw detection · CPC title

• G06V30/42

  based on the type of document · CPC title

• G06V30/18019

  by matching or filtering · CPC title