Image processing method and image processing apparatus for transforming objects within an image

Patent metadata

Field	Value
Publication number	US-12586147-B2
Application number	US-202218263005-A
Country	US
Kind code	B2
Filing date	Jan 4, 2022
Priority date	Feb 7, 2021
Publication date	Mar 24, 2026
Grant date	Mar 24, 2026

Abstract

The present disclosure provides an image processing method and an image processing apparatus. The image processing method including: performing line detection on a boundary of an object in a scene image to obtain a plurality of candidate lines; taking one of the plurality of the candidate lines, which has the largest number of pixels overlapping with the boundary, as a first reference line; generating a minimum bounding rectangle of the object in the scene image; generating a transformed area, wherein each of four vertexes of the minimum bounding rectangle is on one of four sidelines of the transformed area; transforming a preset image into the transformed area to generate a transformed image; and overlaying the object with the transformed image to update the scene image.

First claim

What is claimed is:

1 . An image processing method, comprising:
performing line detection on a boundary of an object in a scene image to obtain a plurality of candidate lines; taking one of the plurality of the candidate lines, which has the largest number of pixels overlapping with the boundary, as a first reference line; generating a minimum bounding rectangle of the object in the scene image; generating a transformed area, wherein each of four vertexes of the minimum bounding rectangle is on one of four sidelines of the transformed area, a slope of a first sideline and a slope of a second sideline opposite to the first sideline of the transformed area are determined by a slope of the first reference line, and a slope of a third sideline and a slope of a fourth sideline opposite to the third sideline of the transformed area are determined by a slope of an exterior angle bisector of the first sideline and the second sideline, wherein the generating the transformed area comprises:
generating the first sideline passing through a first vertex of the minimum bounding rectangle and the second sideline passing through a second vertex of the minimum bounding rectangle, wherein an angle where the first vertex is located and an angle where the second vertex is located are opposite angles, a difference between the slope of the first sideline and the slope of the first reference line and a difference between the slope of the second sideline and the slope of the first reference line are within a first preset range;
generating a second reference line passing through a center point of the minimum bounding rectangle, wherein a slope of the second reference line is the slope of the external angle bisector of the first sideline and the second sideline;
generating the third sideline passing through a third vertex of the minimum bounding rectangle and the fourth sideline passing through a fourth vertex of the minimum bounding rectangle, wherein an angle where the third vertex is located and an angle where the fourth vertex is located are opposite angles, a difference between the slope of the third sideline and the slope of the second reference line and a difference between the slope of the fourth sideline and the slope of the second reference line are within a second preset range; and
generating the transformed area according to the first sideline, the second sideline, the third sideline and the fourth sideline;
transforming a preset image into the transformed area to generate a transformed image; and overlaying the object with the transformed image to update the scene image.

2 . The image processing method according to claim 1 , wherein
the slope of the first sideline is greater than the slope of the first reference line and the slope of the second sideline is less than the slope of the first reference line in a case where the slope of the first reference line is less than 0.

3 . The image processing method according to claim 2 , wherein
the slope of the first sideline is less than the slope of the first reference line and the slope of the second sideline is greater than the slope of the first reference line in a case where the slope of the first reference line is greater than 0.

4 . The image processing method according to claim 3 , wherein:
the difference between the slope of the first sideline and the slope of the first reference line is a ratio of a distance from the first vertex to the first reference line to a first preset parameter; and the difference between the slope of the second sideline and the slope of the first reference line is a ratio of a distance from the second vertex to the first reference line to a second preset parameter.

5 . The image processing method according to claim 4 , wherein
the slope of the third sideline is less than the slope of the second reference line and the slope of the fourth sideline is greater than the slope of the second reference line in a case where the slope of the second reference line is greater than 0.

6 . The image processing method according to claim 5 , wherein
the slope of the third sideline is greater than the slope of the second reference line and the slope of the fourth sideline is less than the slope of the second reference line in a case where the slope of the second reference line is less than 0.

7 . The image processing method according to claim 6 , wherein:
the difference between the slope of the third sideline and the slope of the second reference line is a ratio of a distance from the third vertex to the second reference line to a third preset parameter; and the difference between the slope of the fourth sideline and the slope of the second reference line is a ratio of a distance from the fourth vertex to the second reference line to a fourth preset parameter.

8 . The image processing method according to claim 1 , wherein the transforming a preset image into the transformed area to generate a transformed image comprises:
generating a minimum bounding rectangle of the transformed area; splicing repeatedly the preset image in the minimum bounding rectangle of the transformed area to generate a sample image; and transforming the sample image into the transformed area by affine transformation to generate a transformed image.

9 . The image processing method according to claim 1 , further comprising:
deleting portions of the transformed image beyond the scene image after overlaying the object with the transformed image.

10 . An image processing apparatus, comprising:
a memory; and a processor coupled to the memory, which is configured to execute the image processing method comprising: performing line detection on a boundary of an object in a scene image to obtain a plurality of candidate lines; taking one of the plurality of the candidate lines, which has the largest number of pixels overlapping with the boundary, as a first reference line; generating a minimum bounding rectangle of the object in the scene image; generating a transformed area, wherein each of four vertexes of the minimum bounding rectangle is on one of four sidelines of the transformed area, a slope of a first sideline and a slope of a second sideline opposite to the first sideline of the transformed area are determined by a slope of the first reference line, and a slope of a third sideline and a slope of a fourth sideline opposite to the third sideline of the transformed area are determined by a slope of an exterior angle bisector of the first sideline and the second sideline, wherein the generating the transformed area comprises:
generating the first sideline passing through a first vertex of the minimum bounding rectangle and the second sideline passing through a second vertex of the minimum bounding rectangle, wherein an angle where the first vertex is located and an angle where the second vertex is located are opposite angles, a difference between the slope of the first sideline and the slope of the first reference line and a difference between the slope of the second sideline and the slope of the first reference line are within a first preset range;
generating a second reference line passing through a center point of the minimum bounding rectangle, wherein a slope of the second reference line is the slope of the external angle bisector of the first sideline and the second sideline;
generating the third sideline passing through a third vertex of the minimum bounding rectangle and the fourth sideline passing through a fourth vertex of the minimum bounding rectangle, wherein an angle where the third vertex is located and an angle where the fourth vertex is located are opposite angles, a difference between the slope of the third sideline and the slope of

Assignees

• Beijing Wodong Tianjun Information Technology Co Ltd
• Beijing Jingdong Century Trading Co Ltd
• Beijing Jingoong Century Trading Co Ltd

Inventors

• Yang Mei
• Zhou Fangru
• An Shan

Classifications

• G06T2207/20224

  Image subtraction · CPC title

• G06T7/13

  Edge detection · CPC title

• G06T7/11

  Region-based segmentation · CPC title

• G06T7/187

  involving region growing; involving region merging; involving connected component labelling · CPC title

• G06T11/00

  Two-dimensional [2D] image generation · CPC title