Object detection

Patent metadata

Field	Value
Publication number	US-10706530-B2
Application number	US-201715700684-A
Country	US
Kind code	B2
Filing date	Sep 11, 2017
Priority date	Sep 11, 2017
Publication date	Jul 7, 2020
Grant date	Jul 7, 2020

Abstract

This disclosure provides a method for object detection. The method comprises receiving a user input that specifies one or more first regions and one or more second regions in a template image. The one or more second regions include one or more objects of interest. The method further comprises for each of the one or more first regions discovering a third region in an image under detection corresponding to the first region in the template image by matching the image under detection with the template image. The method further comprises computing a transformation function based on the matching from each of the one or more first regions to its corresponding third region. The method further comprises applying the computed transformation function to the one or more second regions to localize one or more fourth regions in the image under detection for the object detection.

First claim

What is claimed is:

1. A processor-implemented method for object detection, the method comprising:
receiving a user input that specifies a plurality of first regions and one or more second regions in a template image, wherein the one or more second regions include one or more objects of interest;
for each of the plurality of first regions, finding a third region in an image under detection corresponding to one of the plurality of first regions in the template image by matching the image under detection with the template image;
computing a transformation function based on the matching from each of the plurality of first regions to its corresponding third region and a distance between two or more pairs of first regions; and
applying the computed transformation function to the one or more second regions to localize one or more fourth regions in the image under detection for the object detection.

2. The method according to claim 1 , wherein computing a transformation function comprises: computing the transformation function based on the matching from a subset of the first regions to a corresponding subset of the third regions when the number of the first regions are larger than one.

3. The method according to claim 2 , wherein the subset of the first regions is selected based on a similarity between each of the first regions and its matched third region and a distance between each pair of the first regions.

4. The method according to claim 3 , wherein a first region is selected into the subset of the first regions if the first region has a high similarity with its matched third region and exceeds a threshold distance from other first regions.

5. The method according to claim 1 , wherein the one or more first regions are one or more points in the template image, and the third region is one point in the image under detection.

6. The method according to claim 5 , wherein matching is selected from a group consisting of point matching and graph matching.

7. The method according to claim 1 , wherein a first region and a corresponding third region have less variations in appearance compared with a second region and a corresponding fourth region.

8. A computer system for object detection, the computer system comprising:
one or more processors;
one or more computer-readable memories coupled to at least one of the one or more processors;
a set of computer program instructions stored in the memory and executed by at least one of the one or more processors in order to perform actions of:
receiving a user input that specifies a plurality of first regions and one or more second regions in a template image, wherein the one or more second regions include one or more objects of interest;
for each of the plurality of first regions, finding a third region in an image under detection corresponding to one of the plurality of first regions in the template image by matching the image under detection with the template image;
computing a transformation function based on the matching from each of the plurality of first regions to its corresponding third region and a distance between two or more pairs of first regions; and
applying the computed transformation function to the one or more second regions to localize one or more fourth regions in the image under detection for the object detection.

9. The computer system according to claim 8 , wherein computing a transformation function comprising: computing the transformation function based on the matching from a subset of the first regions to a corresponding subset of the third regions when the number of the first regions are larger than one.

10. The computer system according to claim 9 , wherein the subset of the first regions is selected based on a similarity between each of the first regions and its matched third region and a distance between each pair of the first regions.

11. The computer system according to claim 10 , wherein a first region is selected into the subset of the first regions if the first region has a high similarity with its matched third region and exceeds a threshold distance from other first regions.

12. The computer system according to claim 8 , wherein the one or more first regions are one or more points in the template image, and the third region is one point in the image under detection.

13. The computer system according to claim 12 , wherein the matching is selected from a group consisting of point matching and graph matching.

14. The computer system according to claim 8 , wherein a first region and a corresponding third region have less variations in appearance compared with a second region and a corresponding fourth region.

15. A computer program product for object detection, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, wherein the program instructions being executable by a device to cause the device to perform a method comprising:
receiving a user input that specifies a plurality of first regions and one or more second regions in a template image, wherein the one or more second regions include one or more objects of interest;
for each of the plurality of first regions, finding a third region in an image under detection corresponding to one of the plurality of first regions in the template image by matching the image under detection with the template image;
computing a transformation function based on the matching from each of the plurality of first regions to its corresponding third region and a distance between two or more pairs of first regions; and
applying the computed transformation function to the one or more second regions to localize one or more fourth regions in the image under detection for the object detection.

16. The computer program product according to claim 15 , wherein computing a transformation function comprising: computing the transformation function based on the matching from a subset of the first regions to a corresponding subset of the third regions when the number of the first regions are larger than one.

17. The computer program product according to claim 16 , wherein the subset of the first regions is selected based on a similarity between each of the first regions and its matched third region and a distance between each pair of the first regions.

18. The computer program product according to claim 17 , wherein a first region is selected into the subset of the first regions if the first region has a high similarity with its matched third region and exceeds a threshold distance from other first regions.

19. The computer program product according to claim 15 , wherein the one or more first regions are one or more points in the template image, and the third region is one point in the image under detection.

20. The computer program product according to claim 15 , wherein a first region and a corresponding third region have fewer variations in appearance compared with a second region and a corresponding fourth region.

Assignees

• IBM

Inventors

• Hu Guo Qiang
• Huang Jing Chang
• Yan jun chi
• Zhu Jun

Classifications

• G06V10/7788

  the supervisor being a human, e.g. interactive learning with a human teacher · CPC title

• G06V10/774

  Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06V10/25 Primary

  Determination of region of interest [ROI] or a volume of interest [VOI] · CPC title

• G06T7/001 Primary

  using an image reference approach · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title