System, computer program product and associated methodology for video motion detection using spatio-temporal slice processing

• US 8,121,424 B2
• Filed: 11/18/2008
• Issued: 02/21/2012
• Est. Priority Date: 09/26/2008
• Status: Active Grant

• Current Assignee: Axis AB (Canon Inc.)
• Original Assignee: Axis AB (Canon Inc.)
• Inventors: Almbladh, Johan
• Primary Examiner(s): Do, Anh Ohng
• Application Number:
  US12/273,220
• Publication Number:
  US 20100080477A1
• Time in Patent Office: 1,190 Days
• Field of Search: 382/232, 382/240, 382/236, 348/607, 348/699, 375/240.11, 375/E7.04, 375/E7.107, 375/E5.081
• US Class Current: 382/236
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments

Abstract

A system, computer program product and associated methodology for video motion detection combines a series of images of a video sequence into an image volume having vertices of X, Y and t. The image volume is sliced in either the (X,t) or (Y,t) planes, filtered with a spatio-temporal filter and thresholded to reduce the amount of information. Then a search algorithm searches along search lines in a subset of the video sequence to locate motion. Motion can be detected by identifying lines that are not parallel to any of the vertices. Stationary subjects and lighting changes appear as lines parallel to one of the vertices. Thus, true motion can be distinguished from lighting changes.

Claims

1. A video motion detecting method, comprising:

capturing a plurality of images of a video sequence, each of the plurality of images including a plurality of pixels, each pixel being represented by a first number of bits;

reducing the first number of bits to a second number of bits that is less than the first number of bits;

storing in a computer readable medium the plurality of images;

searching with a processor for spatio-temporal edges in a spatio-temporal domain of a predetermined area in the plurality of images; and

detecting motion of a common object in said plurality of images, including

comparing connected strings of pixels have a common value in said predetermined area, and

identifying a presence of an edge when at least one string of connected pixels is greater than a predetermined length.

2. The method of claim 1, wherein

said connected string being arranged in a straight diagonal direction.

3. The method of claim 1, wherein

said connected string being arranged in a predetermined pattern other than a straight diagonal line.

4. The method of claim 1, wherein said reducing step includes reducing the first number of bits to one bit per pixel.

5. The method of claim 4, further comprising a step of packing bits representing respective pixels in a common data word that can be processed by a processor in a single operation instruction.

6. The method of claim 1, wherein said identifying step includes identifying an edge when the at least one string of connected pixels is at least as long as a depth of the predetermined area.

7. The method of claim 1, wherein said search step includes searching adjacent pixels in parallel.

8. The method of claim 1, wherein:

said comparing step includes performing a logical AND operation on sets of pixels in said predetermined area; and

said identifying step includes performing a logical OR operation on connected strings of pixels within said predetermined area.

9. The method of claim 1, wherein said reducing step includes:

performing a logarithmic transformation on the plurality of pixels so as to separate reflectance components from an illumination component.

10. The method of claim 1, wherein the reducing step includes a spatio-temporal filtering step.

11. The method of claim 10, wherein the reducing step includes performing a threshold operation on an absolute value of a result of said spatio-temporal filtering step.

12. A non-transitory computer-readable medium having computer-readable instructions thereon, the computer-readable instructions when executed by a processor cause the processor to perform steps comprising:

capturing a plurality of images of a video sequence, each of the plurality of images including a plurality of pixels, each pixel being represented by a first number of bits;

reducing the first number of bits to a second number of bits that is less than the first number of bits;

storing in a computer readable medium the plurality of images;

searching with the processor for spatio-temporal edges in a spatio-temporal domain of a predetermined area in the plurality of images;

detecting with a processor motion of a common object in said plurality of images, including, said detecting step including

comparing connected strings of pixels have a common value in said predetermined area, and

identifying a presence of an edge when at least one string of connected pixels is greater than a predetermined length.

13. The non-transitory computer-readable medium of claim 12, wherein

said connected string being arranged in a straight diagonal direction.

14. The non-transitory computer-readable medium of claim 12, wherein

said connected string being arranged in a predetermined pattern other than a straight diagonal line.

15. The non-transitory computer-readable medium of claim 12, wherein said reducing step includes reducing the first number of bits to one bit per pixel.

16. The non-transitory computer-readable medium of claim 15, further comprising a step of packing bits representing respective pixels in a common data word that can be processed by a processor in a single operation instruction.

17. The non-transitory computer-readable medium of claim 12, wherein said identifying step includes identifying an edge when the at least one string of connected pixels is at least as long as a depth of the predetermined area.

18. The non-transitory computer-readable medium of claim 12, wherein said search step includes searching adjacent pixels in parallel.

19. The non-transitory computer-readable medium of claim 12, wherein:

said comparing step includes performing a logical AND operation; and

said identifying step includes performing a logical OR operation on connected strings of pixels within said predetermined area.

20. The non-transitory computer-readable medium of claim 12, wherein said reducing step includes:

performing a logarithmic transformation on the plurality of pixels so as to separate reflectance components from an illumination component.

21. The non-transitory computer-readable medium of claim 12, wherein the reducing step includes a spatio-temporal filtering step.

22. The non-transitory computer-readable medium of claim 21, wherein the reducing step includes performing a threshold operation on an absolute value of a result of said spatio-temporal filtering step.

23. A camera comprising:

an optics section configured to capture light from a scene containing an object;

a sensor configured to convert said light into a plurality of images of a video sequence, each of the plurality of images including a plurality of pixels, each pixel being represented by a first number of bits;

an image processor configured to reduce the first number of bits to a second number of bits that is less than the first number of bits;

a memory configured to store the plurality of images with the second number of bits per pixel; and

a processor configured to

search said pixels for spatio-temporal edges in a spatio-temporal domain of a predetermined area in the plurality of images, and

detect motion of the object in said plurality of images by comparing connected strings of pixels have a common value in said predetermined area, and identifying a presence of an edge when at least one string of connected pixels is greater than a predetermined length.

24. The camera of claim 23, wherein

said connected string being arranged in a straight diagonal direction.

25. The camera of claim 23, wherein

said connected string being arranged in a predetermined pattern other than a straight diagonal line.

26. The camera of claim 23, wherein said reducing step includes reducing the first number of bits to one bit per pixel.

27. The camera of claim 26, wherein said image processor is configured to pack bits representing respective pixels in a common data word that can be processed by the processor in a single operation instruction.

28. The camera of claim 23, wherein said processor is configured to identify an edge when the at least one string of connected pixels has a length that is at least as long as a depth of the predetermined area.

29. The camera of claim 23, wherein said search step includes searching adjacent pixels in parallel.

30. The camera of claim 23, wherein:

said processor is configured to perform a logical AND operation on pixels in said predetermined area and perform a logical OR operation on connected strings of pixels within said predetermined area.

31. The camera of claim 23, wherein said image processor is configured to

perform a logarithmic transformation on the plurality of pixels so as to separate a reflectance component from an illumination component.

32. The camera of claim 31, wherein said image processor is also configured to perform spatio-temporal filtering when reducing the first number of bits.

33. The camera of claim 32, wherein said image processor is also configured to perform a threshold operation on an absolute value of a result of said spatio-temporal filtering.