Object recognition device, object recognition method, program for object recognition method, and recording medium having recorded thereon program for object recognition method

• US 8,160,366 B2
• Filed: 06/19/2009
• Issued: 04/17/2012
• Est. Priority Date: 06/20/2008
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Nakamura, Akira, Iwai, Yoshiaki, Yoshigahara, Takayuki
• Primary Examiner(s): Tabatabai, Abolfazl
• Application Number:
  US12/488,099
• Publication Number:
  US 20090316990A1
• Time in Patent Office: 1,033 Days
• Field of Search: 382/100, 382/164, 382/171, 382/173, 382/177, 382/179, 382/190, 382/199, 382/201, 382/218, 702/190, 702/193, 704/221, 704/243, 704/247, 704/231, 705/7.33, 348/94, 348/169, 348/170, 348/171, 348/172, 348/208.14
• US Class Current: 382/190
• CPC Class Codes: G06V 10/462 Salient features, e.g. scale invariant feature transforms [SIFT]
• CPC Class Codes: G06V 10/757 Matching configurations of points or features

Abstract

An object recognition device includes: a model image processing unit having a feature point set decision unit setting a feature point set in a model image, and detecting the feature quantity of the feature point set, and a segmentation unit segmenting the model image; a processing-target image processing unit having a feature point setting unit setting a feature point in a processing-target image and detecting the feature quantity of the feature point; a matching unit comparing the feature quantities of the feature points set in the model image and in the processing-target image so as to detect the feature point corresponding to the feature point set, and executes a matching; and a determination unit determining the processing result in the matching unit so as to determine presence/absence of a model object in the processing-target image.

Claims

1. An object recognition device comprising:

a model image processing unit processing a model image;

a processing-target image processing unit possessing a processing-target image;

a matching unit matching the processing results in the model image processing unit and the processing-target image processing unit; and

a determination unit determining the processing result in the matching unit so as to determine presence/absence of a model object in the processing-target image,

wherein the model image processing unit has a feature point set decision unit setting a feature point set including a base point and a support point supporting the base point in the model image, and detecting the feature quantity of the feature point set,

the processing-target image processing unit has a feature point setting unit setting a feature point in the processing-target image and detecting the feature quantity of the feature point,

the matching unit compares the feature quantity of the feature point set in the model image with the feature quantity of the feature point in the processing-target image so as to detect the feature point corresponding to the feature point set, and executes the matching,

the model image processing unit has a segmentation unit segmenting the model image, and

the feature point set decision unit sets the base point and the corresponding support point in the same segment for each segment of the model image so as to set the feature point set.

2. An object recognition device comprising:

a model image processing unit processing a model image;

a processing-target image processing unit processing a processing-target image;

a matching unit matching the processing results in the model image processing unit and the processing-target image processing unit; and

a determination unit determining the processing result in the matching unit so as to determine presence/absence of a model object in the processing-target image,

wherein the processing-target image processing unit has a feature point set decision unit setting a feature point set including a base point and a support point supporting the base point in the processing-target image, and detecting the feature quantity of the feature point set,

the model image processing unit has a feature point setting unit setting a feature point in the model image and detecting the feature quantity of the feature point,

the matching unit compares the feature quantity of the feature point set in the processing-target image with the feature quantity of the feature point in the model image so as to detect the feature point corresponding to the feature point set, and executes the matching,

the processing-target image processing unit has a segmentation unit segmenting the processing-target image, and

the feature point set decision unit sets the base point and the corresponding support point in the same segment for each segment of the processing-target image so as to set the feature point set.

3. An object recognition method comprising the steps of:

processing a model image;

processing a processing-target image;

matching the processing results in the step of processing the model image and the step of processing the processing-target image; and

determining the processing result in the step of matching so as to determine presence/absence of a model object in the processing-target image,

wherein the step of processing the model image has a substep of setting a feature point set including a base point and a support point supporting the base point in the model image, and detecting the feature quantity of the feature point set, thereby deciding the feature point set,

the step of processing the processing-target image has a substep of setting a feature point in the processing-target image and detecting the feature quantity of the feature point,

in the step of matching, the feature point corresponding to the feature point set is detected by comparison of the feature quantity of the feature point set in the model image with the feature quantity of the feature point in the processing-target image, and the matching is executed,

the step of processing the model image has a substep of segmenting the model image, and

in the step of deciding the feature point set, the base point and the corresponding support point are set in the same segment for each segment of the model image so as to set the feature point set.

4. The method according to claim 3,

wherein in the step of deciding the feature point set, the base point and the support point are set on a segment boundary.

5. The method according to claim 3,

wherein in the step of deciding the feature point set, the number of feature point sets to be set in one segment is set on the basis of the area and peripheral length of the corresponding segment.

6. The method according to claim 3,

wherein in the step of deciding the feature point set, the number of support points constituting one feature point set is set on the basis of the area and peripheral length of the corresponding segment.

7. The method according to claim 3,

wherein the step of deciding the feature point set has a substep of setting the base point and/or the support point on a segment boundary, and setting the feature quantity description center of the base point and/or the support point at a position displaced inward of the corresponding segment from the base point and/or the support point,

a local area of the base point and/or the support point is set with the feature quantity description center as a center, and

the feature quantity of the base point and/or the support point is detected on the basis of the feature quantity of the local area so as to set the feature quantity of the feature point set.

8. The method according to claim 7,

wherein the substep of setting the feature quantity description center has substeps of

shrinking a processing-target segment to the size corresponding to the local area,

thinning the processing-target segment, and

superimposing an area generated in the step of shrinking on an area thinned in the step of thinning so as to set an area where the feature quantity description center is set,

the feature quantity description center is set in the outer circumference of the area where the feature quantity description center is set, and

in the step of deciding the feature point set, the base point and/or the support point is set on the basis of the feature quantity description center.

9. The method according to claim 3,

wherein in the step of setting the feature point, all pixels of the processing-target image are set as feature points, or feature points are set in the processing-target image with pitches corresponding to a predetermined number of pixels,

a local area is set for each feature point, and the feature quantity of the local area is detected, and

in the step of matching, the feature point corresponding to the feature point set is detected by detection of the feature point corresponding to the base point and the support point of the feature point set in the model image, and the matching is executed.

10. The method according to claim 9,

wherein in the step of setting the feature point, a feature point on the edge of the processing-target image is selected and set as a processing-target feature point of the step of matching.

11. The method according to claim 9,

wherein the step of determining includes the substeps of

determining the processing result in the step of matching so as to primarily determine presence/absence of a model object in the processing-target image,

setting all pixels of the model image as feature points or setting feature points in the model image with pitches corresponding to a predetermined number of pixels in accordance with the determination result in the substep of primarily determining, and detecting the feature quantity of each feature point of the model image, thereby reprocessing the model image,

setting the feature point set in the processing-target image in accordance with the determination result in the substep of primarily determining, thereby reprocessing the processing-target image,

executing back matching by comparison of the feature quantity of the feature point set in the processing-target image by the step of reprocessing the processing-target image with the feature quantity of the feature point in the model image by the step of reprocessing the model image, and

determining the processing result in the step of back matching so as to secondarily determine presence/absence of a model object in the processing-target image.

12. An object recognition method comprising the steps of:

processing a model image;

processing a processing-target image;

matching the processing results in the step of processing the model image and the step of processing the processing-target image; and

determining the processing result in the step of matching so as to determine presence/absence of a model object in the processing-target image,

wherein the step of processing the processing-target image has a substep of setting a feature point set including a base point and a support point supporting the base point in the processing-target image, and detecting the feature quantity of the feature point set, thereby deciding the feature point set,

the step of processing the model image has a substep of setting a feature point in the model image and detecting the feature quantity of the feature point,

in the step of matching, the feature point corresponding to the feature point set is detected by comparison of the feature quantity of the feature point set in the processing-target image and the feature quantity of the feature point in the model image, and the matching is executed,

the step of processing the processing-target image has a substep of segmenting the processing-target image, and

in the step of deciding the feature point set, the base point and the corresponding support point are set in the same segment for each segment of the processing-target image so as to set the feature point set.

13. The method according to claim 12,

wherein in the step of deciding the feature point set, the base point and the support point are set on a segment boundary.

14. The method according to claim 12,

wherein in the step of deciding the feature point set, the number of feature point sets to be set in one segment is set on the basis of the area and peripheral length of the corresponding segment.

15. The method according to claim 12,

wherein in the step of deciding the feature point set, the number of support points constituting one feature point set is set on the basis of the area and peripheral length of the corresponding segment.

16. The method according to claim 12,

wherein the step of deciding the feature point set has a substep of setting the base point on a segment boundary, and setting the feature quantity description center of the base point at a position displaced inward of the corresponding segment from the base point,

a local area of the base point is set with the feature quantity description center as a center, and

the feature quantity of the base point is detected on the basis of the feature quantity of the local area so as to set the feature quantity of the feature point set.

17. The method according to claim 16,

wherein the substep of setting the feature quantity description center has substeps of

shrinking a processing-target segment to the size corresponding to the local area,

thinning the processing-target segment, and

superimposing an area generated in the step of shrinking on an area thinned in the step of thinning so as to set an area where the feature quantity description center is set,

the feature quantity description center is set in the outer circumference of the area where the feature quantity description center is set, and

in the step of deciding the feature point set, the corresponding base point is set on the basis of the feature quantity description center.

18. The method according to claim 12,

wherein in the step of setting the feature point, all pixels of the model image are set as feature points, or feature points are set in the model image with pitches corresponding to a predetermined number of pixels,

a local area is set for each feature point, and the feature quantity of the local area is detected, and

in the step of matching, the feature point corresponding to the feature point set is detected by detection of the feature point corresponding to the base point and the support point of the feature point set in the processing-target image, and the matching is executed.

19. The method according to claim 18,

wherein in the step of setting the feature point, a feature point on the edge of the processing-target image is selected and set as a processing-target feature point of the step of matching.

20. The method according to claim 18,

wherein the step of determining includes the substeps of

determining the processing result in the step of matching so as to primarily determine presence/absence of a model object in the processing-target image,

setting all pixels of the processing-target image as feature points or setting feature points in the processing-target image with pitches corresponding to a predetermined number of pixels in accordance with the determination result in the substep of primarily determining, and detecting the feature quantity of each feature point of the processing-target image, thereby reprocessing the processing-target image,

setting the feature point set in the model image in accordance with the determination result in the substep of primarily determining, thereby reprocessing the model image,

executing back matching by comparison of the feature quantity of the feature point set in the model image by the step of reprocessing the model image with the feature quantity of the feature point in the processing-target image by the step of reprocessing the processing-target image, and

determining the processing result in the step of back matching so as to secondarily determine presence/absence of a model object in the processing-target image.

21. A non-transitory computer readable medium having a program stored thereon for an object recognition method that is executable by a computer, the program comprising the steps of:

processing a model image;

processing a processing-target image;

matching the processing results in the step of processing the model image and the step of processing the processing-target image; and

determining the processing result in the step of matching so as to determine presence/absence of a model object in the processing-target image,

wherein the step of processing the model image has a substep of setting a feature point set including a base point and a support point supporting the base point in the model image, and detecting the feature quantity of the feature point set, thereby deciding the feature point set,

the step of processing the processing-target image has a substep of setting a feature point in the processing-target image and detecting the feature quantity of the feature point,

in the step of matching, the feature point corresponding to the feature point set is detected by comparison of the feature quantity of the feature point set in the model image with the feature quantity of the feature point in the processing-target image, and the matching is executed,

the step of processing the model image has a substep of segmenting the model image, and

in the step of deciding the feature point set, the base point and the corresponding support point are set in the same segment for each segment of the model image so as to set the feature point set.

22. A non-transitory computer readable medium having a program stored thereon for an object recognition method that is executable by a computer, the program comprising the steps of:

processing a model image;

processing a processing-target image;

matching the processing results in the step of processing the model image and the step of processing the processing-target image; and

determining the processing result in the step of matching so as to determine presence/absence of a model object in the processing-target image;

wherein the step of processing the processing-target image has a substep of setting a feature point set including a base point and a support point supporting the base point in the processing-target image, and detecting the feature quantity of the feature point set, thereby deciding the feature point set,

the step of processing the model image has a substep of setting a feature point in the model image and detecting the feature quantity of the feature point,

in the step of matching, the feature point corresponding to the feature point set is detected by comparison of the feature quantity of the feature point set in the processing-target image and the feature quantity of the feature point in the model image, and the matching is executed,

the step of processing the processing-target image has a substep of segmenting the processing-target image,

in the step of deciding the feature point set, the base point and the corresponding support point are set in the same segment for each segment of the processing-target image so as to set the feature point set.

23. A non-transitory recording medium having recorded thereon a program for an object recognition method, which is executable by a computer, the program comprising the steps of:

processing a model image;

processing a processing-target image;

matching the processing results in the step of processing the model image and the step of processing the processing-target image; and

determining the processing result in the step of matching so as to determine presence/absence of a model object in the processing-target image,

wherein the step of processing the model image has a substep of setting a feature point set including a base point and a support point supporting the base point in the model image, and detecting the feature quantity of the feature point set, thereby deciding the feature point set,

the step of processing the processing-target image has a substep of setting a feature point in the processing-target image and detecting the feature quantity of the feature point,

in the step of matching, the feature point corresponding to the feature point set is detected by comparison of the feature quantity of the feature point set in the model image with the feature quantity of the feature point in the processing-target image, and the matching is executed,

the step of processing the model image has a substep of segmenting the model image, and

in the step of deciding the feature point set, the base point and the corresponding support point are set in the same segment for each segment of the model image so as to set the feature point set.

24. A non-transitory recording medium having recorded thereon a program for an object recognition method, which is executable by a computer, the program comprising the steps of:

processing a model image;

processing a processing-target image;

matching the processing results in the step of processing the model image and the step of processing the processing-target image; and

determining the processing result in the step of matching so as to determine presence/absence of a model object in the processing-target image,

wherein the step of processing the processing-target image has a substep of setting a feature point set including a base point and a support point supporting the base point in the processing-target image, and detecting the feature quantity of the feature point set, thereby deciding the feature point set,

the step of processing the model image has a substep of setting a feature point in the model image and detecting the feature quantity of the feature point,

in the step of matching, the feature point corresponding to the feature point set is detected by comparison of the feature quantity of the feature point set in the processing-target image and the feature quantity of the feature point in the model image, and the matching is executed,

the step of processing the processing-target image has a substep of segmenting the processing-target image,

in the step of deciding the feature point set, the base point and the corresponding support point are set in the same segment for each segment of the processing-target image so as to set the feature point set.