Efficient detection of camera shake

• US 8,094,959 B2
• Filed: 02/14/2008
• Issued: 01/10/2012
• Est. Priority Date: 12/09/2005
• Status: Active Grant

• Current Assignee: Seiko Epson Corporation (Seiko Group)
• Original Assignee: Seiko Epson Corporation (Seiko Group)
• Inventors: Rai, Barinder Singh
• Primary Examiner(s): PATEL, KANJIBHAI B
• Application Number:
  US12/030,964
• Publication Number:
  US 20080144955A1
• Time in Patent Office: 1,426 Days
• Field of Search: 382/173, 382/232, 382254-255, 382274-276, 382/284, 382/305, 382/312, 396/52, 348/208.4, 348/208.99
• US Class Current: 382/254
• CPC Class Codes: G06V 10/56 relating to colour
• CPC Class Codes: H04N 1/00307 with a mobile telephone apparatus constructional features of portable telephone sets H04M1/0202; mobile radio systems H04W84/00
• CPC Class Codes: H04N 1/00352 Input means H04N1/00411 takes precedence; input arrangements for computers G06F3/00
• CPC Class Codes: H04N 1/00381 Input by recognition or interpretation of visible user gestures manual input means, e.g. digitisers, writing tablets H04N1/00392; gesture input to computers G06F3/017; recognition algorithms G06V40/20
• CPC Class Codes: H04N 1/00395 Arrangements for reducing operator input input arrangements for job or mode instructions or parameters H04N1/00962; intelligent menu interfaces H04N1/00437
• CPC Class Codes: H04N 2101/00 Still video cameras
• CPC Class Codes: H04N 2201/0084 Digital still camera

Abstract

At least two sites in a frame of pixels are specified. The sites are arranged in a particular spatial distribution and correspond with the pixel locations of a block of pixels. Block parameters are calculated for each pixel block of first and second frames. The block parameters may be calculated using fewer than all of the bits of each pixel. A block-pair similarity determination for each pair of spatially-corresponding pixel blocks of the first and second frames is generated by determining whether there is a difference between the respective block parameters which is greater than a particular block-level threshold. A frame similarity determination is generated by combining the block-pair similarity determinations. A user-interface indication may be provided, or a frame may be stored, as a result of the frame similarity determination.

Claims

1. A method comprising:

using a display controller to;

(a) specify at least two sites in a frame, the sites being arranged in a particular spatial distribution, each site corresponding with the pixel locations of a block of pixels, each block having at least two pixels;

(b) calculate a block parameter for each block of first and second frames in a sequence of frames, each block parameter being calculated from the pixels of a respective one of the blocks;

(c) generate a block-pair similarity determination for each pair of spatially-corresponding pixel blocks of the first and second frames by determining whether there is a difference between the respective block parameters which is greater than a particular block-level threshold; and

(d) generate a frame similarity determination by combining the block-pair similarity determinations to detect a camera shake when the first and second frames are not similar and to indicate a shake-free image is not probable, and to indicate a shake-free image is probable when the first and second frames are similar.

2. The method of claim 1, further comprising specifying nine regions by dividing the frame into three equal-width;

vertical bands and three equal-height, horizontal bands, and setting each intersection of one of the vertical and one of the horizontal bands as one of the regions, wherein the particular spatial distribution situates one site in each of the nine regions.

3. The method of claim 1, further comprising specifying nine regions by dividing the frame into three equal-width, vertical bands and three equal-height, horizontal bands, and setting each intersection of one of the vertical and one of the horizontal bands as one of the regions, wherein a center region is specified by the intersection of a central vertical band and a central horizontal band, top-center and bottom-center regions are respectively specified by the intersections of the central vertical band and top and bottom horizontal bands, and left-center and right-center regions are respectively specified by the intersections of the central horizontal band and left and right vertical bands, wherein the particular spatial distribution situates at least two sites in the center region and one site in each of the top-center, bottom-center, left-center, and right-center regions.

4. The method of claim 1, further comprising specifying nine regions by dividing the frame into three equal-width, vertical bands and three equal-height, horizontal bands, and setting each intersection of one of the vertical and one of the horizontal bands as one of the regions, wherein a center region is specified by the intersection of a central vertical band and a central horizontal band, top-center and bottom-center regions are respectively specified by the intersections of the central vertical band and top and bottom horizontal bands, and left-center and right-center regions are respectively specified by the intersections of the central horizontal band and left and right vertical bands, wherein the particular spatial distribution situates one site in each of the center, top-center, bottom-center, left-center, and right-center regions, and one site centered at each four-corner junction where corners of four regions meet.

5. The method of claim 1, wherein each pixel is defined by at least two bits and the block parameters are calculated using fewer than all of the bits of each pixel.

6. The method of claim 5, further comprising specifying nine regions by dividing the frame into three equal-width, vertical bands and three equal-height, horizontal bands, and setting each intersection of one of the vertical and one of the horizontal bands as one of the regions, wherein the particular spatial distribution situates one site in each of the nine regions.

7. The method of claim 5, further comprising specifying nine regions by dividing the frame into three equal-width, vertical bands and three equal-height, horizontal bands, and setting each intersection of one of the vertical and one of the horizontal bands as one of the regions, wherein a center region is specified by the intersection of a central vertical band and a central horizontal band, top-center and bottom-center regions are respectively specified by the intersections of the central vertical band and top and bottom horizontal bands, and left-center and right-center regions are respectively specified by the intersections of the central horizontal band and left and right vertical bands, wherein the particular spatial distribution situates at least two sites in the center region and one site in each of the top-center, bottom-center, left-center, and right-center regions.

8. The method of claim 5, further comprising specifying nine regions by dividing the frame into three equal-width, vertical bands and three equal-height, horizontal bands, and setting each intersection of one of the vertical and one of the horizontal bands as one of the regions, wherein a center region is specified by the intersection of a central vertical band and a central horizontal band, top-center and bottom-center regions are respectively specified by the intersections of the central vertical band and top and bottom horizontal bands, and left-center and right-center regions are respectively specified by the intersections of the central horizontal band and left and right vertical bands, wherein the particular spatial distribution situates one site in each of the center, top-center, bottom-center, left-center, and right-center regions, and one site centered at each four-corner junction where corners of four regions meet.

9. The method of claim 1, wherein the particular block-level threshold used to generate a block-pair similarity determination for a first pair of spatially-corresponding blocks is distinct from the block-level threshold used to generate a block-pair similarity determination for a second pair of spatially-corresponding blocks.

10. The method of claim 1, wherein generating a frame similarity determination by combining block-pair similarity determinations includes assigning a first weight to a block-pair similarity determination for a first block pair and assigning a second weight to a block-pair similarity determination for a second block pair, the first and second weights being distinct.

11. The method of claim 1, further comprising rendering an indication corresponding with the frame similarity determination in a user-interface device.

12. The method of claim 1, further comprising storing a third frame in the sequence of frames in a nonvolatile memory if the frame similarity determination indicates that a shake-free image is probable.

13. The method of claim 1, further comprising storing the second frame in a nonvolatile memory if the frame similarity determination indicates that a shake-free image is probable.

14. A circuit coupled with a bus for transmitting frames of pixels, the circuit comprising:

a first unit to sample a pixel transmitted on the bus if the coordinate location of the pixel is within one of at least two specified blocks of pixels of a frame, and to calculate a block parameter for each block from the sampled pixels, wherein each of the blocks is situated at a distinct site in the frame and the sites are arranged in a particular spatial distribution;

a first memory to store the block parameters calculated for a first frame in a sequence of frames;

a second unit to generate a block-pair similarity determination for each pair of spatially-corresponding pixel blocks of the first and a second frame in the sequence of frames; and

a third unit to a generate frame similarity determination by combining the block-pair similarity determinations generated for the first and second frames to detect a camera shake when the first and second frames are not similar and to indicate a shake-free image is not probable, and to indicate a shake-free image is probable when the first and second frames are similar.

15. The circuit of claim 14, further comprising a second memory to store a frame of pixels and a fourth unit to control the storing of frames in the second memory, wherein when the fourth unit causes a third frame to be stored in the memory in response to a frame similarity determination which indicates that the first and second frames are similar.

16. The circuit of claim 14, further comprising a fifth unit to cause a user interface device to render an indication corresponding with the frame similarity determination.

17. The circuit of claim 14, wherein each pixel is defined by at least two bits and the block parameters are calculated using fewer than all of the bits of each pixel.

18. The circuit of claim 14, wherein frame similarity determinations are generated in real-time.

19. A system, comprising:

a camera module to output pixels within specified pixel blocks, the specified pixel blocks being situated at distinct sites in a frame, the sites being arranged in a particular spatial distribution, the pixels corresponding with a particular capture event; and

a display controller to (a) calculate a block parameter for each of the specified pixel blocks, (b) generate a block-pair similarity determination for each pair of spatially-corresponding pixel blocks of a first and a second set of pixel blocks, the first and second sets corresponding with first and second capture events, and (c) generate a frame similarity determination by combining the block-pair similarity determinations generated for the first and second sets of pixel blocks to detect a camera shake when the first and second frames are not similar and to indicate a shake-free image is not probable, and to indicate a shake-free image is probable when the first and second frames are similar.

20. The system of claim 19, wherein each pixel is defined by at least two bits and the block parameters are calculated using fewer than all of the bits of each pixel.

21. The system of claim 19, wherein, in response to a particular frame similarity determination, the display controller causes the camera module to output a frame, the frame corresponding with a third capture event.

22. The system of claim 21, further comprising a memory, wherein the display controller stores the frame corresponding with the third capture event in the memory.

23. The system of claim 19, further comprising a display device wherein, in response to a particular frame similarity determination, the display controller causes an indication to be rendered on the display device corresponding with the frame similarity determination.

24. The system of claim 21, wherein the camera module outputs a sequence of frames, each frame in the sequence corresponding with a particular capture event, and the pixels within the specified pixel blocks are included in the sequence of frames, and wherein the display controller identifies pixels within the specified pixel blocks for use in calculating the block parameters.

25. The system of claim 24, wherein, in response to a particular frame similarity determination, the display controller causes the camera module to output a frame, the frame corresponding with a third capture event.

26. The system of claim 25, further comprising a memory, wherein the display controller stores the frame corresponding with a third capture event in the memory.

27. The system of claim 24, wherein, in response to a particular frame similarity determination, the display controller causes the camera module to output a frame, the frame corresponding with the second capture event.

28. The system of claim 27, further comprising a memory, wherein the display controller stores the frame corresponding with the second capture event in the memory.