Method and apparatus for parallelization of image compression encoders

• US 20080247653A1
• Filed: 04/03/2007
• Published: 10/09/2008
• Est. Priority Date: 04/03/2007
• Status: Active Grant

• Current Assignee: Aptina Imaging Corporation (ON Semiconductor Corporation)
• Original Assignee: Micron Technology, Inc.
• Inventors: Lin, Sheng, Moussavi, Farshid, Kopet, Thomas, Jabbi, Aman
• Granted Patent: US 8,098,941 B2
• Time in Patent Office: Days
• US Class Current: 382/232
• CPC Class Codes: H04N 19/186 the unit being a colour or a chrominance component
• CPC Class Codes: H04N 19/436 using parallelised computational arrangements
• CPC Class Codes: H04N 19/60 using transform coding

Abstract

A method and apparatus for compressing image data include a plurality of core encoders for receiving a respective plurality of streams of data in parallel and for compressing the received data, a plurality of output buffers for respectively receiving the compressed data from each of the plurality of core encoders, and a recombination circuit for recombining the compressed data.

Claims

1. An apparatus for compressing image data comprising:

a plurality of core encoders for receiving a respective plurality of streams of data in parallel and for compressing the received data;

a plurality of output buffers for respectively receiving the compressed data from each of the plurality of core encoders; and

a recombination circuit for recombining the compressed data.

2. The apparatus of claim 1, wherein the plurality of core encoders compresses the data according to a JPEG (Joint Photographic Experts Group) standard.

3. The apparatus of claim 1, wherein each of the plurality of streams of data comprises at least one minimum coded unit.

4. The apparatus of claim 3, wherein the at least one minimum coded unit comprises four sets of pixel image data, the sets comprising first and second luminance values, blue chrominance values, and red chrominance values.

5. The apparatus of claim 1, wherein each of the plurality of streams of data comprises a plurality of minimum coded units.

6. The apparatus of claim 5, wherein each of the plurality of streams of data further comprises at least one restart marker.

7. The apparatus of claim 6, wherein the plurality of minimum coded units are separated by the at least one restart marker into a plurality of sets of equal numbers of minimum coded units.

8. The apparatus of claim 1, wherein each of the plurality of streams of data comprises one of luminance values and chrominance values.

9. The apparatus of claim 8, wherein:

at least a first of the plurality of core encoders is configured to receive the luminance values; and

at least a second of the plurality of core encoders is configured to receive the chrominance values.

10. The apparatus of claim 8, wherein:

the luminance values comprise a first set of luminance values and a second set of luminance values; and

the chrominance values comprise blue chrominance values and red chrominance values.

11. The apparatus of claim 10, further comprising:

a plurality of differential encoders for respectively receiving an output of the plurality of output buffers and for outputting differentially encoded data to a respective plurality of pre-recombination buffers; and

at least one Huffman table coupled to the plurality of differential encoders,

wherein the plurality of pre-recombination buffers outputs to the recombination circuit.

12. An apparatus for compressing image data comprising:

a data receiving circuit for receiving a serial stream of image data and for converting the serial stream of image data into a plurality of parallel streams of data; and

the apparatus of claim 1.

14. The apparatus of claim 12, wherein:

each of the plurality of streams of data comprises at least one minimum coded unit, and

the at least one minimum coded unit comprises four sets of pixel image data, the sets comprising first and second luminance values, blue chrominance values, and red chrominance values.

16. The apparatus of claim 12, wherein:

each of the plurality of streams of data comprises a plurality of minimum coded units,

each of the plurality of streams of data further comprises at least one restart marker,

the plurality of minimum coded units are separated by the at least one restart marker into a plurality of sets of equal numbers of minimum coded units, and

the data receiving circuit converts the serial stream of image data into a plurality of parallel streams of data based on the at least one restart marker.

13. (canceled)

15. (canceled)

17-23. -23. (canceled)

24. A method of compressing image data comprising:

respectively receiving, in parallel, a plurality of streams of data;

compressing, in parallel, the received data;

respectively buffering the compressed data from each of the plurality of core encoders; and

recombining the compressed data.

26. The method of claim 24, wherein:

each of the plurality of streams of data comprises at least one minimum coded unit, and

the at least one minimum coded unit comprises four sets of pixel image data, the sets comprising first and second luminance values, blue chrominance values. and red chrominance values.

28. The method of claim 24, wherein:

each of the plurality of streams of data comprises a plurality of minimum coded units,

each of the plurality of streams of data further comprises at least one restart marker,

at least one restart marker separates the plurality of minimum coded units into a plurality of sets of equal numbers of minimum coded units.

31. The method of claim 24, wherein each of the plurality of streams of data respectively comprises one of luminance values and chrominance values.

32. The method of claim 31, further comprising:

receiving the luminance values with at least a first of a plurality of core encoders; and

receiving the chrominance values with at least a second of the plurality of core encoders.

34. The method of claim 31, further comprising:

respectively outputting the compressed data to a plurality of output buffers;

respectively outputting the buffered compressed data to a plurality of differential encoders;

differentially encoding the compressed data using at least one Huffman table coupled to the plurality of differential encoders;

respectively outputting the differentially encoded data to a plurality of pre-recombination buffers; and

outputting the pre-recombination buffered differentially encoded data to a recombination circuit,

wherein;

the luminance values comprise a first set of luminance values and a second set of luminance values; and

the chrominance values comprise blue chrominance values and red chrominance values.

25. (canceled)

27. (canceled)

29-30. -30. (canceled)

33. (canceled)

35. A method of compressing image data comprising:

receiving a serial stream of image data;

converting the serial stream of image data into a plurality of parallel streams of data;

compressing, in parallel, the plurality of parallel streams of data;

respectively buffering the compressed data from each of the plurality of core encoders; and

recombining the compressed data.

46. The method of claim 35, further comprising:

respectively outputting the compressed data to a plurality of output buffers;

respectively outputting the buffered compressed data to a plurality of differential encoders;

differentially encoding the compressed data using at least one Huffman table coupled to the plurality of differential encoders;

respectively outputting the differentially encoded data to a plurality of pre-recombination buffers; and

outputting the pre-recombination buffered differentially encoded data to a recombination circuit,

wherein;

the luminance values comprise a first set of luminance values and a second set of luminance values; and

the chrominance values comprise blue chrominance values and red chrominance values.

36-45. -45. (canceled)

47. An imaging system comprising:

a pixel array and readout circuit for capturing image data; and

a device for compressing captured image data, the device comprising;

a plurality of core encoders for receiving a respective plurality of streams of data in parallel and for compressing the received data;

a plurality of output buffers for respectively receiving the compressed data from each of the plurality of core encoders; and

a recombination circuit for recombining the compressed data.

49. A camera system comprising:

a lens for focusing an image to be captured; and

the imaging system of claim 47.

48. An imaging system comprising:

a pixel array and readout circuit for capturing image data; and

a device for compressing captured image data, the device comprising;

a data receiving circuit for receiving a serial stream of image data and for converting the serial stream of image data into a plurality of parallel streams of data;

a plurality of core encoders for receiving a respective one of the plurality of parallel streams of data and for compressing the respective one of the plurality of parallel streams of data;

a plurality of output buffers for respectively receiving the compressed data from each of the plurality of core encoders; and

a recombination circuit for recombining the compressed data.

50. (canceled)