Photographing apparatus and method for controlling target tracking

• US 8,050,455 B2
• Filed: 06/04/2007
• Issued: 11/01/2011
• Est. Priority Date: 10/25/2006
• Status: Active Grant

• Current Assignee: Hanwha Vision Company Limited (Hanwha Corporation)
• Original Assignee: Samsung Techwin Co., Ltd. (Hanwha Corporation)
• Inventors: Seo, Jin-keun, Lee, Suk-ho, Woo, Hyun-gun, Noh, Seung-in
• Primary Examiner(s): Repko; Jason M
• Assistant Examiner(s): NAKHJAVAN, SHERVIN K
• Application Number:
  US11/806,751
• Publication Number:
  US 20080101655A1
• Time in Patent Office: 1,611 Days
• Field of Search: 382/103, 382/173, 348/169, 352/39
• US Class Current: 382/103
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G06V 10/62 relating to a temporal dimension, e.g. time-based feature extraction; Pattern tracking

Abstract

A photographing apparatus and a method for controlling target tracking are provided. The photographing apparatus includes a target extractor for extracting target information in a present frame based on the difference between a pixel value of one pixel among previous pixels constituting the target of the previous frame in an area adjacent to the present pixel constituting the present frame and a pixel value of the present pixel, and a tracking controller for controlling automatic tracking for the target based on the target information. Accordingly, the target can be extracted more precisely.

Claims

1. A photographing apparatus comprising:

a processor;

a target extractor which extracts target information about a target in a present frame based on information regarding a difference between a first pixel value of a first pixel of the present frame and each of respective first pixel values of a plurality of target pixels, constituting the target in a previous frame, in an area adjacent to the first pixel of the present frame; and

a tracking controller which controls automatic tracking of the target based on the target information,

wherein the target extractor comprises a target-difference image generator which generates a target-difference image by modifying a second pixel value of the first pixel of the present frame to a value corresponding to a difference between the first pixel value of the first pixel and a first pixel value, among the respective first pixel values of the plurality of target pixels, which is the most similar to the first pixel value of the first pixel of the present frame, and

wherein the target extractor extracts the target information in the present frame based on the target-difference image.

2. The photographing apparatus of claim 1, wherein the target-difference image generator generates the target-difference image by modifying the second pixel value of the first pixel to a preset value, if the target in the previous frame is not in the area adjacent to the first pixel.

3. The photographing apparatus of claim 1, wherein the target extractor further comprises a target-difference image divider which generates a target segment of the target by dividing an area of the target-difference image based on second pixel values of pixels constituting the target-difference image, and

wherein the target extractor extracts the target information in the present frame based on the target segment.

4. The photographing apparatus of claim 3, wherein the target-difference image divider generates the target segment comprising pixels having a second pixel value less than a first value from the target-difference image.

5. The photographing apparatus of claim 4, wherein the target-difference image divider regenerates the target segment comprising pixels having a second pixel value less than a second value from the pixels constituting the target segment.

6. The photographing apparatus of claim 1,

wherein the target extractor comprises a background-difference image generator which generates a background-difference image by modifying a first pixel value of a second pixel of the present frame to a difference between the first pixel value of the second pixel and a first pixel value, among respective first pixel values of a plurality of background pixels constituting a background in the previous frame in an area adjacent to the second pixel, which is the most similar to the first pixel value of the second pixel of the present frame, and

wherein the target extractor extracts the target information in the present frame based on the target-difference image and the background-difference image.

7. The photographing apparatus of claim 6, wherein the target extractor further comprises:

a target-difference image divider which generates a target segment of the target by dividing an area of the target-difference image based on second pixel values of pixels constituting the target-difference image;

a background-difference image divider which generates a background segment of the background by dividing an area of the background-difference image based on second pixel values of pixels constituting the background-difference image; and

a target reconstructor which outputs the target information by excluding the pixels constituting the background segment from the pixels constituting the target segment.

8. The photographing apparatus of claim 1, wherein the difference is obtained with respect to substantially all first pixels, including the first pixel, of the present frame.

9. A method for controlling target tracking using a photographing apparatus comprising a processor, the method comprising:

extracting, under control of the processor, target information about a target in a present frame based on information regarding a difference between a first pixel value of a first pixel of the present frame and each of respective first pixel values of a plurality of target pixels, constituting the target in a previous frame, in an area adjacent to the first pixel of the present frame; and

automatically tracking the target based on the target information,

wherein the extracting the target information comprises generating a target-difference image by modifying a second pixel value of the first pixel of the present frame to a value corresponding to a difference between the first pixel value of the first pixel and a first pixel value, among the respective first pixel values of the plurality of target pixels, which is the most similar to the first pixel value of the first pixel of the present frame, and

wherein, in the extracting the target information, the target information in the present frame is extracted based on the target-difference image.

10. The method of claim 9, wherein in the generating the target-difference image, the target-difference image is generated by modifying the second pixel value of the first pixel to a preset value, if the target in the previous frame is not in the area adjacent to the first pixel.

11. The method of claim 9, wherein the extracting the target information further comprises generating a target segment of the target by dividing an area of the target-difference image based on second pixel values of pixels constituting the target-difference image, and

wherein in the extracting the target information, the target information in the present frame is extracted based on the target segment.

12. The method of claim 11, wherein in the generating the target segment of the target, the target segment is generated by dividing pixels having a second pixel value less than a first value from the target-difference image.

13. The method of claim 12, wherein in the generating the target segment of the target, the target segment is regenerated by dividing pixels having a second pixel value less than a second value from the pixels constituting the target segment.

14. The method of claim 9, wherein extracting the target information comprises generating a background-difference image by modifying a first pixel value of a second pixel of the present frame to a difference between the first pixel value of the second pixel and a first pixel value, among respective first pixel values of a plurality of background pixels constituting a background in the previous frame in an area adjacent to the second pixel, which is the most similar to the first pixel value of the second pixel of the present frame, and

wherein, in extracting the target information, the target information in the present frame is extracted based on the target-difference image and the background-difference image.

15. The method of claim 14, wherein the extracting the target information further comprises:

generating a target segment of the target by dividing an area of the target-difference image based on second pixel values of pixels constituting the target-difference image;

generating a background segment of the background by dividing an area of the background-difference image based on second pixel values of pixels constituting the background-difference image; and

outputting the target information by excluding the pixels constituting the background segment from the pixels constituting the target segment.

16. The method of claim 9, wherein the difference is obtained with respect to substantially all first pixels, including the first pixel, of the present frame.

17. A photographing apparatus comprising:

a processor;

a target extractor which, under control of the processor, extracts target information about a target in a present frame based on a similarity between a first pixel value of a first pixel of the present frame and each of respective first pixel values of a plurality of target pixels constituting the target in a previous frame, in an area adjacent to the first pixel of the present frame; and

a tracking controller which controls automatic tracking of the target based on the target information,

wherein the target extractor comprises a first divider which generates a target segment of the target by dividing an area of the present frame based on the similarity,

wherein the target extractor extracts the target information based on the target segment

wherein the target extractor further comprises a second divider which generates a background segment of a background by dividing another area of the present frame based on a similarity between a first pixel value of a second pixel of the present frame and each of respective first pixel values of a plurality of background pixels constituting the background in the previous frame in an area adjacent to the second pixel; and

wherein the target extractor extracts the target information based on the target segment and the background segment.

18. The photographing apparatus of claim 17, wherein the target extractor further comprises a target reconstructor which outputs the target information by excluding pixels constituting the background segment from pixels constituting the target segment.

19. A method for controlling target tracking using a photographing apparatus comprising a processor, the method comprising:

extracting, under control of the processor, target information about a target in a present frame based on a similarity between a first pixel value of a first pixel of the present frame and each of respective first pixel values of a plurality of target pixels constituting the target in a previous frame, in an area adjacent to the first pixel of the present frame; and

controlling automatic tracking of the target based on the target information,

wherein extracting the target information comprises generating a target segment of the target by dividing an area of the present frame based on the similarity,

wherein the target information is extracted based on the target segment

wherein extracting the target information further comprises generating a background segment of a background by dividing another area of the present frame based on a similarity between a first pixel value of a second pixel of the present frame and each of respective first pixel values of a plurality of background pixels constituting the background in the previous frame in an area adjacent to the second pixel; and

wherein the target information is extracted based on the target segment and the background segment.

20. The method of claim 19, wherein extracting the target information further comprises outputting the target information by excluding pixels constituting the background segment from pixels constituting the target segment.

21. A non-transitory computer-readable medium having embodied thereon a computer program for executing a method of controlling target tracking, the method comprising:

extracting, under control of the processor, target information about a target in a present frame based on information regarding a difference between a first pixel value of a first pixel of the present frame and each of respective first pixel values of a plurality of target pixels, constituting the target in a previous frame, in an area adjacent to the first pixel of the present frame; and

automatically tracking the target based on the target information,

wherein the extracting the target information comprises generating a target-difference image by modifying a second pixel value of the first pixel of the present frame to a value corresponding to a difference between the first pixel value of the first pixel and a first pixel value, among the respective first pixel values of the plurality of target pixels, which is the most similar to the first pixel value of the first pixel of the present frame, and

wherein, in the extracting the target information, the target information in the present frame is extracted based on the target-difference image.

22. A non-transitory computer-readable medium having embodied thereon a computer program for executing the method of controlling target tracking, the method comprising:

extracting, under control of the processor, target information about a target in a present frame based on a similarity between a first pixel value of a first pixel of the present frame and each of respective first pixel values of a plurality of target pixels constituting the target in a previous frame, in an area adjacent to the first pixel of the present frame; and

controlling automatic tracking of the target based on the target information,

wherein the extracting the target information comprises generating a target segment of the target by dividing an area of the present frame based on the similarity,

wherein the target information is extracted based on the target segment,

wherein extracting the target information further comprises generating a background segment of a background by dividing another area of the present frame based on a similarity between a first pixel value of a second pixel of the present frame and each of respective first pixel values of a plurality of background pixels constituting the background in the previous frame in an area adjacent to the second pixel; and

wherein the target information is extracted based on the target segment and the background segment.