Method of image processing

• US 9,424,486 B2
• Filed: 04/16/2014
• Issued: 08/23/2016
• Est. Priority Date: 08/16/2006
• Status: Active Grant

• Current Assignee: Zebra Technologies Corporation
• Original Assignee: Cortexica Vision Systems Limited (Zebra Technologies Corporation)
• Inventors: Bharath, Anil Anthony, Ng Sing Kwong, Jeffrey
• Primary Examiner(s): Bali, Vikkram
• Application Number:
  US14/254,851
• Publication Number:
  US 20140226859A1
• Time in Patent Office: 860 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06T 7/74 involving reference images or patches
• CPC Class Codes: G06V 10/42 Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation
• CPC Class Codes: G06V 40/168 Feature extraction; Face representation

Abstract

Disclosed embodiments relate to a systems and methods of identifying a target object in an image using image processing. Template data is stored which represents a template orientation field indicative of an orientation of each of a plurality of features of a template object. Image data may be received and processed to generate an image orientation field indicating an orientation corresponding to the plurality of image features. The image data may be further processed to generate image confidence data based on at least one characteristic for use in identifying the target object in a given image, and the characteristic may be indicative of an increased likelihood that at least one part of the given image represents at least part of the target object, relative to other parts of the given image. The image orientation field is processed using the template orientation field and the image confidence data to generate a match metric indicative of an extent of matching between at least part of the template orientation field and at least part of the image orientation field. The match metric may then be used to determine whether or not the target object has been identified in the image.

Claims

1. A method of identifying a target object in an image including a plurality of image features, the method comprising:

storing template data, the template data representing a template orientation field indicative of an orientation of each of a plurality of features of a template object;

receiving image data representing the image;

processing the image data to generate an image orientation field indicating an orientation corresponding to the plurality of image features;

processing the image data to generate image confidence data based on at least one characteristic for use in identifying the target object in a given image, the characteristic being indicative of an increased likelihood that at least one part of the given image represents at least part of the target object, relative to other parts of the given image;

processing the image orientation field using the template orientation field and the image confidence data to generate a match metric indicative of an extent of matching between at least part of the template orientation field and at least part of the image orientation field; and

using the match metric to determine whether or not the target object has been identified in the image.

2. The method of claim 1, the method comprises applying the image confidence data to the image orientation field to generate a modified image orientation field indicating both an orientation corresponding to the plurality of image features and a likelihood that at least one part of the image orientation field represents at least part of the target object.

3. The method of claim 1, wherein the characteristic indicated by the image confidence data relates to an extent of image data errors expected in the image data, and wherein the method further comprises using the image confidence data to generate the match metric according to the expected extent of image data errors.

4. The method of claim 1, wherein the characteristic indicated by the image confidence data relates to a possible occlusion of part of the target object in the image, and wherein the method further comprises using the image confidence data to generate the match metric according to the possible occlusion.

5. The method of claim 1, wherein the image data is video image data and wherein the characteristic relates to a possible movement of the target object.

6. The method of claim 5, wherein the method further comprises using the characteristic relating to the possible movement of the target object and an identified movement of the target object to determine a modified shape of the target object, and wherein the target object is modified from a shape of the target object represented by the template data.

7. The method of claim 1, wherein the image data represents the image in color, and wherein the characteristic relates to at least one color indicative of the target object.

8. The method of claim 1, wherein generating the match metric comprises using deformation data indicative of a deformation of template data with respect to image data.

9. The method of claim 1, wherein at least one of the image data, the template data, the image confidence data, and deformation data used to generate the match metric, is arranged according to a matrix format.

10. The method of claim 1, wherein each of the image features represents, in the image, a boundary which separates a region of higher light intensity from a region of lower light intensity.

11. The method of claim 10, wherein the boundary separates two regions, each having a similar light intensity, from each other.

12. A system for identifying a target object in an image including a plurality of image features, the system comprising:

a data store comprising executable software;

a processor in data communication with the data store, the processor configured to execute the software and cause a computing device to;

store template data, said template data representing a template orientation field indicative of an orientation of each of a plurality of features of a template object;

receive image data representing the image;

process the image data to generate an image orientation field indicating an orientation corresponding to the plurality of image features;

process the image data to generate image confidence data based on at least one characteristic for use in identifying the target object in a given image, the characteristic being indicative of an increased likelihood that at least one part of the given image represents at least part of the target object, relative to other parts of the given image;

process the image orientation field using the template orientation field and the image confidence data to generate a match metric indicative of an extent of matching between at least part of the template orientation field and at least part of the image orientation field; and

use the match metric to determine whether or not the target object has been identified in the image.

13. A non-transitory computer-readable storage medium having computer readable instructions stored thereon, wherein, when executed, cause the computerized device to perform a method of identifying a target object in an image including a plurality of image features, the method comprising, the method comprising:

storing template data, said template data representing a template orientation field indicative of an orientation of each of a plurality of features of a template object;

receiving image data representing the image;

processing the image data to generate an image orientation field indicating an orientation corresponding to the plurality of image features;

processing the image data to generate image confidence data based on at least one characteristic for use in identifying the target object in a given image, the characteristic being indicative of an increased likelihood that at least one part of the given image represents at least part of the target object, relative to other parts of the given image;

processing the image orientation field using the template orientation field and the image confidence data to generate a match metric indicative of an extent of matching between at least part of the template orientation field and at least part of the image orientation field; and

using the match metric to determine whether or not the target object has been identified in the image.

14. A method of identifying a target object in an image including a plurality of image features, the method comprising:

storing template data, the template data representing a template orientation field indicative of an orientation of each of a plurality of features of a template object;

receiving template confidence data indicating a likelihood that at least one part of the template accurately represents at least part of the target object;

receiving image data representing the image;

processing the image data to generate an image orientation field indicating an orientation corresponding to the plurality of image features;

processing the image orientation field using the template orientation field and the template confidence data to generate a match metric indicative of an extent of matching between at least part of the template orientation field and at least part of the image orientation field; and

using the match metric to determine whether or not the target object has been identified in the image.

15. The method of claim 14, wherein the method further comprises applying the template confidence data to the template orientation field to generate a modified template orientation field indicating both an orientation of each of the plurality of features of the template object and a likelihood that at least one part of the template orientation field represents at least part of the target object.

16. The method of claim 14, wherein generating the match metric comprises using deformation data indicative of a deformation of template data with respect to image data.

17. The method of claim 14, wherein each of the image features represents a boundary in the image which separates a region of higher light intensity from a region of lower light intensity.

18. The method of claim 17, wherein the boundary separates two regions, each having a similar light intensity, from each other.

19. A system for identifying a target object in an image including a plurality of image features, the system comprising:

a data store comprising executable software;

a processor in data communication with the data store, the processor configured to execute the software and cause a computing device to;

store template data, the template data representing a template orientation field indicative of an orientation of each of a plurality of features of a template object;

receive template confidence data indicating a likelihood that at least one part of the template accurately represents at least part of the target object;

receive image data representing the image;

process the image data to generate an image orientation field indicating an orientation corresponding to the plurality of image features;

process the image orientation field using the template orientation field and the template confidence data to generate a match metric indicative of an extent of matching between at least part of the template orientation field and at least part of the image orientation field; and

use the match metric to determine whether or not the target object has been identified in the image.

20. A non-transitory computer-readable storage medium having computer readable instructions stored thereon, wherein, when executed, cause the computerized device to perform a method of identifying a target object in an image including a plurality of image features, the method comprising:

storing template data, the template data representing a template orientation field indicative of an orientation of each of a plurality of features of a template object;

receiving template confidence data indicating a likelihood that at least one part of the template accurately represents at least part of the target object;

receiving image data representing the image;

processing the image data to generate an image orientation field indicating an orientation corresponding to the plurality of image features;

processing the image orientation field using the template orientation field and the template confidence data to generate a match metric indicative of an extent of matching between at least part of the template orientation field and at least part of the image orientation field; and

using the match metric to determine whether or not the target object has been identified in the image.