IMAGE PROCESSING APPARATUS AND METHOD

• US 20070122006A1
• Filed: 09/14/2006
• Published: 05/31/2007
• Est. Priority Date: 09/16/2005
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: TANIGUCHI, Hidenori
• Granted Patent: US 8,194,935 B2
• Time in Patent Office: Days
• US Class Current: 382/116
• CPC Class Codes: G06V 10/75 Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
• CPC Class Codes: G06V 40/173 face re-identification, e.g. recognising unknown faces across different face tracks

Abstract

Plural kinds of feature information for identifying an object and a period of time, set for each kind of the feature information, for which the feature information is effective are stored in connection with each other. A feature is extracted from image data, the extracted feature is compared with the feature information, which is within the set period of time, among the stored feature information, and an object is recognized.

Claims

1. An image processing apparatus comprising:

a memory that stores plural kinds of feature information for identifying an object and a period of time, set for each kind of the feature information, for which the feature information is effective in connection with each other;

an extraction unit that extracts a feature from image data; and

a matching unit that compares the feature extracted by said extraction unit with the feature information, which is within the set period of time, among the feature information stored in said memory and recognizes an object.

2. The image processing apparatus according to claim 1, further comprising:

a judgment unit that judges whether the set period of time of the feature information stored in said memory has elapsed; and

a deletion unit that deletes the feature information whose set period of time has elapsed from the memory.

3. The image processing apparatus according to claim 1, further comprising:

a judgment unit that judges whether the set period of time of the feature information stored in said memory has elapsed; and

a warning unit that warns expiration of the feature information whose set period of time has elapsed.

4. The image processing apparatus according to claim 1, wherein the feature information includes a long term feature which is set with a period of time for which the feature information is effective and a short term feature which is set with a period of time for which the feature information is effective being shorter than that of the long term feature.

5. The image processing apparatus according to claim 4, wherein said memory stores features classified into the long term feature and the short term feature.

6. The image processing apparatus according to claim 1, wherein the feature information includes a long term feature which is not set with a period of time for which the feature information is effective and a short term feature which is set with a period of time for which the feature information is effective.

7. The image processing apparatus according to claim 6, wherein said memory stores features classified into the long term feature and the short term feature.

8. The image processing apparatus according to claim 6, wherein said matching unit gives a priority to the long term feature upon comparing a feature extracted by said extraction unit with feature information, and performs comparison with the short term feature if an object could not be recognized by comparing with the long term feature.

9. The image processing apparatus according to claim 1, further comprising:

an image sensing unit that performs photoelectric conversion of light from an object to output electrical image data; and

an image sensing condition controller that controls an image sensing condition used in said image sensing unit based on an recognition result of an object by said matching unit.

10. An image processing method comprising:

a storing step of storing plural kinds of feature information for identifying an object and a period of time, set for each kind of the feature information, for which the feature information is effective in connection with each other;

an extraction step of extracting a feature from image data; and

a matching step of comparing the feature extracted in said extraction step with the feature information, which is within the set period of time, among the feature information stored in said storing step and recognizing an object.

11. The image processing method according to claim 10, further comprising:

a judging step of judging whether the set period of time of the feature information stored in said storing step has elapsed; and

a deletion step of deleting the feature information whose set period of time has elapsed.

12. The image processing method according to claim 10, further comprising:

a judging step of judging whether the set period of time of the feature information stored in said storing step has elapsed; and

a warning step of warning expiration of the feature information whose set period of time has elapsed.

13. The image processing method according to claim 10, wherein the feature information includes a long term feature which is set with a period of time for which the feature information is effective and a short term feature which is set with a period of time for which the feature information is effective being shorter than that of the long term feature.

14. The image processing method according to claim 13, wherein, in said storing step, features are classified into the long term feature and the short term feature.

15. The image processing method according to claim 10, wherein the feature information includes a long term feature which is not set with a period of time for which the feature information is effective and a short term feature which is set with a period of time for which the feature information is effective.

16. The image processing method according to claim 15, wherein, in said storing step, features are classified into the long term feature and the short term feature.

17. The image processing method according to claim 15, wherein, in said matching step, a priority is given to the long term feature upon comparing a feature extracted in said extraction step with the feature information, and comparison with the short term feature is performed if an object could not be recognized by comparing with the long term feature.

18. The image processing method according to claim 10, further comprising:

an image sensing step of performing photoelectric conversion of light from an object to output electrical image data; and

a step of controlling an image sensing condition in said image sensing step based on an recognition result of an object by said matching step.

19. A storage medium readable by an information processing apparatus, said storage medium storing a program which is executable by the information processing apparatus and comprises program codes realizing the image processing method described in claim 10.