Digital image processor for color image transmission

• US 5,909,254 A
• Filed: 09/08/1993
• Issued: 06/01/1999
• Est. Priority Date: 07/30/1992
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Feig, Ephraim, Linzer, Elliot
• Primary Examiner(s): Harvey, David E.
• Application Number:
  US08/117,971
• Time in Patent Office: 2,092 Days
• Field of Search: 358/133, 358/13, 358/30, 358/11, 358/12, 358/15, 358/40, 348/659-661, 348/713, 348/720
• US Class Current: 348/660
• CPC Class Codes: H04N 11/044 involving transform coding
• CPC Class Codes: H04N 9/67 for matrixing camera processing pipelines for matrixing of colour signals H04N23/85
• CPC Class Codes: H04N 9/8047 using transform coding

Abstract

A method and apparatus for increasing the processing speed or decreasing circuit complexity (or both) of a digital image processor by eliminating the conventional multiplication operation from the RGB to YUV and YUV to RGB transforms having NTSC compatible equation coefficients. In particular, the change of color-basis matrix for converting RGB data to obtain YUV data is factored into a scale factor and a coefficient matrix. The coefficient matrix contains coefficients such that its product with the RGB input vector can be performed using only seven binary additions and five shift operations. This process effectively removes the conventional multiplication operations from the RGB to YUV conversion. The scale factor is conveniently absorbed into the quantization of the YUV data before transmission. The quantization process includes a multiplication already and the two multiplication steps are folded into one step. Therefore, the absorption of the scale factor into the quantization step does not increase the processing time because no multiplication steps are added. This process of factoring the transform matrix is also applied to the inverse transform process. In the inverse transform process the change of color-basis matrix for converting YUV data to RGB data is such that its implementation reduced to five addition steps and two shift steps, again effectively eliminating the conventional multiplication operations. Also, the diagonal matrix is absorbed into the descaling process so that no multiplication steps are added to the inverse transform process.

Claims

1. A method for transmitting an image over a data channel, comprising:

organizing digital RGB data which represents said image by pixel location;

transforming said digital RGB data into digital YUV data which represents said image through an RGB to YUV transform by creating RGB signals corresponding to said digital RGB data, storing said RGB signals in storage locations, creating YUV signals corresponding to said digital YUV data by shifting and adding said RGB signals within said storage locations;

filtering said YUV signals to reduce the quantity of said storage locations containing said filtered YUV signals corresponding to said digital YUV data representing said image;

scaling said filtered YUV signals by adjusting the magnitude of said filtered YUV signals and storing said adjusted magnitudes in storage locations;

converting said scaled YUV signals for transmission over said data channel; and

applying said signals to said data channel.

2. A method for transmitting an image over a data channel, as in claim 1, wherein:

said YUV signals are filtered and subsampled to reduce the quantity of said filtered YUV signals representing said image.

3. A method for transmitting an image over a data channel, as in claim 1, wherein:

a scale factor multiplied by a factored matrix forms said RGB to YUV transform; and

said scaling step includes multiplying by said scale factor.

4. A method for transmitting an image over a data channel, comprising:

organizing digital RGB data which represents said image by pixel location;

transforming said digital RGB data into digital YUV data which represents said image wherein Y=2.5R+5G+B, U=2.5R+5G +7.5B, and V=-6R+5B+B by creating RGB signals corresponding to said digital RGB data, storing said RGB signals in storage locations creating YUV signals corresponding to said digital YUV data by shifting and adding parts of said RGB signals within said storage locations;

filtering said YUV signals to reduce the quantity of said storage locations containing said YUV signals corresponding to said YUV digital data representing said image;

scaling said filtered YUV signals by adjusting the magnitude of said filtered YUV signals;

converting said scaled YUV signals into signals for transmission over said data channel; and

applying said signals to said data channel.

5. A method for transmitting an image over a data channel, as in claim 4, wherein:

said YUV signals are filtered and subsampled to reduce the quantity of said YUV signals representing said image.

6. A method for transmitting an image over a data channel, as in claim 4, wherein:

said scaling process absorbs a constant having a value of approximately 0.118.

7. A method for transmitting an image over a data channel, as in claim 4, wherein:

said scaling process includes multiplying by approximately ##EQU13##

8. A method for transmitting an image over a data channel, as in claim 6, wherein:

said RGB to YUV transformation process is composed of right shifting R data one binary position;

adding G data to said shifted R data to obtain an intermediate value λ

1;

left shifting said λ

1 value two binary positions;

adding said left shifted λ

1 value to said λ

1 value to obtain an intermediate value λ

2;

adding said λ

2 value to B data to obtain a value value Y;

left shifting said R data three binary positions;

adding said left shifted R data to said right shifted R data to obtain an intermediate value λ

3;

adding said Y value to a negative λ

3 to obtain a value U;

left shifting said B data three binary positions;

right shifting said B data one binary position;

adding said left shifted B data to said right shifted B data to obtain an intermediate value λ

4; and

adding said Y value to a negative λ

4 to obtain a value V.

9. A method for transmitting an image over a data channel, as in claim 7, wherein:

said RGB to YUV transformation process is composed of right shifting R data one binary position;

adding G data to said shifted R data to obtain an intermediate value λ

1;

left shifting said λ

1 value two binary positions;

adding said left shifted λ

1 value to said λ

1 value to obtain an intermediate value λ

2;

adding said λ

2 value to B data to obtain a value value Y;

left shifting said R data three binary positions;

adding said left shifted R data to said right shifted R data to obtain an intermediate value λ

3;

adding said Y value to a negative λ

3 to obtain a value U;

left shifting said B data three binary positions;

right shifting said B data one binary position;

adding said left shifted B data to said right shifted B data to obtain an intermediate value λ

4; and

adding said Y value to a negative λ

4 to obtain a value V.

10. A method for displaying an image, comprising:

sensing signals from a data channel;

decoding said sensed signals into YUV signals corresponding to digital YUV data and storing said YUV signals in storage locations;

descaling said stored YUV signals by adjusting the magnitude of said stored YUV signals and storing said adjusted magnitudes in storage locations;

transforming said digital YUV data into digital RGB data through a YUV to RGB transform by creating YUV signals corresponding to said digital YUV data, storing said YUV signals in storage locations, creating RGB signals corresponding to said digital RGB data by shifting and adding said YUV signals within storage locations; and

displaying said RGB data on a display means.

11. A method for displaying an image, as in claim 10, wherein:

a diagonal matrix multiplied by a factored matrix forms said YUV to RGB transform; and

said descaling step includes multiplying by said diagonal matrix.

12. A method for displaying an image, comprising:

sensing signals of digital YUV data from a data channel;

decoding said sensed signals into YUV signals corresponding to digital YUV data and storing said YUV digital signals in storage locations;

descaling said digital YUV data;

transforming said digital YUV data into RGB digital data through a YUV to RGB transform wherein R=Y-V, G=Y+U+0.5V, and B=Y-5U by creating YUV signals corresponding to said digital YUV data, storing said YUV signals in storage locations, creating digital RGB data by shifting and adding said YUV signals within storage locations; and

displaying said digital RGB data on a display means.

13. A method for displaying an image, as in claim 12 wherein;

said descaling process includes multiplying by ##EQU14##

14. A method for displaying an image, as in claim 12 wherein;

said descaling process includes multiplying by ##EQU15##

15. A method of displaying an image, as in claim 13, wherein;

said YUV to RGB transform is composed of right shifting said V data one binary position;

adding said U data to said right shifted V data to obtain an intermediate value λ

1;

left shifting said U data two binary positions;

adding said U data to said left shifted U data to obtain an intermediate value λ

2;

adding said Y data to negative V data to obtain a R value;

adding said Y data to said λ

1 to obtain a G value; and

adding said Y data to a negative λ

2 to obtain a B value.

16. A method of displaying an image, as in claim 14, wherein;

said YUV to RGB transform is composed of right shifting said V data one binary position;

adding said U data to said right shifted V data to obtain an intermediate value λ

1;

left shifting said U data two binary positions;

adding said U data to said left shifted U data to obtain an intermediate value λ

2;

adding said Y data to negative V data to obtain a R value;

adding said Y data to said λ

1 to obtain a G value; and

adding said Y data to a negative λ

2 to obtain a B value.

17. A transmitter for transmitting an image over a data channel, comprising:

a transformer means for transforming digital RGB data into digital YUV data which represents said image by creating RGB signals corresponding to said digital RGB data, storing said RGB signals in storage locations, creating YUV signals corresponding to said digital YUV data by shifting and adding parts of said RGB signals within said storage locations;

a filter means for filtering said YUV signals to reduce the quantity of said storage locations containing said YUV signals corresponding to said digital YUV data representing said image;

a scaling means for scaling said filtered YUV signals by adjusting the magnitude of said filtered YUV signals and storing said adjusted magnitudes in storage locations;

a conversion means for converting said scaled YUV signals into signals for transmission over said data channel; and

a driver means for applying said signals to said data channel.

18. A transmitter for transmitting an image over a data channel, as in claim 17, wherein:

said UV data is filtered and subsampled to reduce the quantity of data representing said image.

19. A transmitter for transmitting an image over a data channel, as in claim 17, wherein:

said RGB to YUV transformation includes a scale factor multiplied by a factored matrix;

said transformer means multiplies said RGB data by said factored matrix; and

said scaling means includes multiplying by said scale factor.

20. A transmitter for transmitting an image over a data channel, comprising:

a transformer means for transforming digital RGB data into digital YUV data which represents said image wherein said transformer means generates Y=2.5R+5G+B,U=2.5R+5G-7.5 B, and V=-6R+5G+B by creating RGB signals corresponding to said digital RGB data, storing said RGB signals in storage locations, creating YUV signals corresponding to said digital YUV data by shifting and adding parts of said RGB signals within said storage locations;

filter means for creating filtered YUV signals by filtering said YUV signals to reduce the quantity of data representing said image;

scaling means for scaling said filtered YUV signals to reduce the magnitude of said filtered YUV signals to create scaled YUV signals

conversion means for converting said scaled YUV signals into signals for transmission over said data channel; and

driver means for applying said signals to said data channel.

21. A transmitter for transmitting an image over a data channel, as in claim 20, wherein:

said UV data is filtered and subsampled to reduce the quantity of data representing said image.

22. A transmitter for transmitting an image over a data channel, as in claim 20, wherein:

said scaling means includes multiplying by scale factor of approximately 0.118.

23. A transmitter for transmitting an image over a data channel, as in claim 20, wherein:

said scaling means includes multiplying by ##EQU16##

24. A transmitter for transmitting an image over a data channel, as in claim 22, wherein:

said transform means right shifts R data one binary position;

said transform means adds G data to said shifted R data to obtain an intermediate value λ

1;

said transform means left shifts said λ

1 value two binary positions;

said transform means adds said left shifted λ

1 value to said λ

1 value to obtain an intermediate value λ

2;

said transform means adds said λ

2 value to B data to obtain a final value value Y;

said transform means left shifts said R data three binary positions;

said transform means adds said left shifted R data to said right shifted R data to obtain an intermediate value λ

3;

said transform means adds said Y value to a negative λ

3 to obtain a final value U;

said transform means left shifts said B data three binary positions;

said transform means right shifts said B data one binary position;

said transform means adds said left shifted B data to said right shifted B data to obtain an intermediate value λ

4; and

said transform means adds said Y value to a negative λ

4 to obtain a final value V.

25. A transmitter for transmitting an image over a data channel, as in claim 23, wherein:

said transform means right shifts R data one binary position;

said transform means adds G data to said shifted R data to obtain an intermediate value λ

1;

said transform means left shifts said λ

1 value two binary positions;

said transform means adds said left shifted λ

1 value to said λ

1 value to obtain an intermediate value λ

2;

said transform means adds said λ

2 value to B data to obtain a final value value Y;

said transform means left shifts said R data three binary positions;

said transform means adds said left shifted R data to said right shifted R data to obtain an intermediate value λ

3;

said transform means adds said Y value to a negative λ

3 to obtain a final value U;

said transform means left shifts said B data three binary positions;

said transform means right shifts said B data one binary position;

said transform means adds said left shifted B data to said right shifted B data to obtain an intermediate value λ

4; and

said transform means adds said Y value to a negative λ

4 to obtain a final value V.

26. A display system for displaying an image, comprising:

a sensor means for sensing signals from a data channel;

decoding means for decoding said sensed signals into digital YUV data and storing said digital YUV data in storage locations;

a descaling means for descaling said digital YUV data by adjusting the magnitude of said digital YUV data and storing said adjusted magnitudes in storage locations;

a transform means for transforming said digital YUV data into RGB digital data by creating YUV signals corresponding to said digital YUV data, storing said YUV signals in storage locations, creating RGB digital data by shift and adding said YUV signals within said storage locations; and

display means for displaying said RGB data.

27. A display system for displaying an image, as in claim 26, wherein:

said YUV to RGB transformation multiplies a diagonal matrix by a factored matrix;

said transform means multiplies said YUV data by said factored matrix; and

said descaling means includes multiplying by said diagonal matrix.

28. A display system for displaying an image, comprising:

sensor means for sensing signals from a data channel;

decoding means for decoding said sensed signals into digital YUV data and storing said digital YUV data in storage locations;

descaling means for descaling said digital YUV data;

transform means for transforming said digital YUV data into digital RGB data wherein R=Y-V, G=Y=U+0.5V, and B=Y=U, by creating YUV signals corresponding to said digital YUV data, storing said YUV signals in storage locations, creating RGB signals corresponding to said digital RGB data by shifting and adding said YUV signals within said storage locations; and

display means for displaying said digital RGB data.

29. A display system for displaying an image, as in claim 28 wherein;

said descaling means includes multiplying by ##EQU17##30.

30. A display system for displaying an image, as in claim 28 wherein;

said descaling means includes multiplying by ##EQU18##

31. A display means for displaying an image, as in claim 29, wherein;

said transform means right shirts said V data one binary position;

said transform means adds said U data to said right shifted V data to obtain an intermediate value λ

1;

said transform means left shifts said U data two binary positions;

said transform means adds said U data to said left shifted U data to obtain an intermediate value λ

2;

said transform means adds said Y data to negative V data to obtain a final R value;

said transform means adds said Y data to said λ

1 to obtain a final G value; and

said transform means adds said Y data to a negative λ

2 to obtain a final B value.

32. A display system for displaying an image, as in claim 30, wherein;

said transform means right shifts said V data one binary position;

said transform means adds said U data to said right shifted V data to obtain an intermediate value λ

1;

said transform means left shifts said U data two binary positions;

said transform means adds said U data to said left shifted U data to obtain an intermediate value λ

2;

said transform means adds said Y data to negative V data to obtain a final R value;

said transform means adds said Y data to said λ

1 to obtain a final G value; and

said transform means adds said Y data to a negative λ

2 to obtain a final B value.

33. A method for storing an image, comprising:

organizing digital RGB data which represents said image by pixel location;

transforming said digital RGB data into digital YUV data which represents said image through an RGB to YUV transform by creating RGB signals corresponding to said digital RGB data, storing said RGB signals in storage locations, creating YUV signals corresponding to said digital YUV data by shifting and adding parts of said RGB signals within said storage locations;

filtering said YUV signals to create filtered YUV signals to reduce the quantity of said storage locations containing YUV signals corresponding to said digital YUV data representing said image;

scaling said filtered YUV signals by adjusting the magnitude of said filtered YUV signals and storing said adjusted magnitudes in storage locations; and

applying said signals to a signal storage means.

34. A method for storing an image, as in claim 33, wherein:

said UV data is filtered and subsampled to reduce the quantity of data representing said image.

35. A method for storing an image, as in claim 33, wherein:

a scale factor multiplied by a factored matrix forms said RGB to YUV transform; and

said scaling step includes multiplying by said scale factor.

36. A method for storing an image, comprising:

organizing digital RGB data which represents said image by pixel location;

transforming said digital RGB data into digital YUV data which represents said image wherein Y=-6R+5G+B by creating RGB signals corresponding to said digital RGB data, storing said RGB signals in storage locations, creating YUV signals corresponding to said digital YUV data by shifting and adding parts of said RGB signals within said storage locations;

filtering said YUV signals to create filtered YUV signals to reduce the quantity of said storage locations containing said YUV signals corresponding to said digital YUV data representing said image;

scaling said filtered YUV signals by adjusting the magnitude of said filtered YUV signals and storing said adjusted magnitudes in storage locations;

converting said scaled YUV signals into signals to be stored; and

applying said signals to a storage means.

37. A method for storing an image, as in claim 36, wherein:

said UV data is filtered and subsampled to reduce the quantity of data representing said image.

38. A method for storing an image, as in claim 36, wherein:

said scaling process includes multiplying by a constant having a value of approximately 0.118.

39. A method for storing an image, as in claim 36, wherein:

said scaling process includes multiplying by ##EQU19##40.

40. A method for storing an image, as in claim 38, wherein:

said RGB to YUV transformation process is composed of right shifting R data one binary position;

adding G data to said shifted R data to obtain an intermediate value λ

1;

left shifting said λ

1 value two binary positions;

adding said left shifted λ

1 value to said λ

1 value to obtain an intermediate value λ

2;

adding said λ

2 value to B data to obtain a final value value Y;

left shifting said R data three binary positions;

adding said left shifted R data to said right shifted R data to obtain an intermediate value λ

3;

adding said Y value to a negative λ

3 to obtain a final value U;

left shifting said B data three binary positions;

right shifting said B data one binary position;

adding said left shifted B data to said right shifted B data to obtain an intermediate value λ

4; and

adding said Y value to a negative λ

4 to obtain a final value V.

41. A method for storing an image, as in claim 39, wherein:

said RGB to YUV transformation process is composed of right shifting R data one binary position;

adding G data to said shifted R data to obtain an intermediate value λ

1;

left shifting said λ

1 value two binary positions;

adding said left shifted λ

1 value to said λ

1 value to obtain an intermediate value λ

2;

adding said λ

2 value to B data to obtain a value value Y;

left shifting said R data three binary positions;

adding said left shifted R data to said right shifted R data to obtain an intermediate value λ

3;

adding said Y value to a negative λ

3 to obtain a value U;

left shifting said B data three binary positions;

right shifting said B data one binary position;

adding said left shifted B data to said right shifted B data to obtain an intermediate value λ

4; and

adding said Y value to a negative λ

4 to obtain a value V.

42. A storage system for storing an image, comprising:

a transformer means for transforming digital RGB into digital YUV data which represents said image by creating RGB signals corresponding to said digital RGB data, storing said RGB signals in storage locations, creating YUV signals corresponding to said digital RGB data, storing said RGB signals in storage locations, creating YUV signals corresponding to said digital YUV data by shifting and adding parts of said RGB signals within said storage locations;

a filter means for filtering said YUV signals to create filtered YUV signals to reduce the quantity of said storage locations containing said YUV signals corresponding to said digital YUV data representing said image;

a scaling means for scaling said filtered YUV signals by adjusting the magnitude of said filtered YUV signals and storing said adjusted magnitude in storage locations;

a conversion means for converting said scaled YUV signals into signals for storage; and

a signal storage means for storing said signals.

43. A storage system for storing an image, as in claim 42, wherein:

said UV data is filtered and subsampled to reduce the quantity of data representing said image.

44. A storage system for storing an image, as in claim 42, wherein:

said RGB to YUV transformation includes a scale factor multiplied by a factored matrix;

said transformer means multiplies said RGB data by said factored matrix; and

said scaling means includes multiplying by said scale factor.

45. A storage system for storing an image, comprising:

a transformer means for transforming digital RGB data into digital YUV data which represents said image wherein said transformer means generates Y=2.5R+5G+B,U=2.5R+5G-7.5 B, and V=-6R+5G+B by creating RGB signals corresponding to said digital RGB data, storing said RGB signals in storage locations, creating YUV signals corresponding to said digital YUV data by shifting and adding parts of said RGB signals within said storage locations;

filter means for filtering said YUV signals to create filtered YUV signals to reduce the quantity of said storage locations containing said YUV signals corresponding to said digital YUV data representing said image;

scaling means for scaling said filtered YUV signals by adjusting the magnitude of said filtered YUV signals and for storing said adjusting magnitudes in storage locations;

conversion means for converting said scaled data into signals for storage; and

a signal storage means for storing said signals.

46. A storage system for storing an image, as in claim 45, wherein:

said UV data is filtered and subsampled to reduce the quantity of data representing said image.

47. A storage system for storing an image, as in claim 45, wherein:

said scaling means includes multiplying by a scale factor of approximately 0.118.

48. A storage system for storing an image, as in claim 45, wherein:

said scaling means includes multiplying by ##EQU20##

49. A storage system for storing an image, as in claim 47, wherein:

said transform means right shifts R data one binary position;

said transform means adds G data to said shifted R data to obtain an intermediate value λ

1;

said transform means left shifts said λ

1 value two binary positions;

said transform means adds said left shifted λ

1 value to said λ

1 value to obtain an intermediate value λ

2;

said transform means adds said λ

2 value to B data to obtain a value value Y;

said transform means left shifts said R data three binary positions;

said transform means adds said left shifted R data to said right shifted R data to obtain an intermediate value λ

3;

said transform means adds said Y value to a negative λ

3 to obtain a value U;

said transform means left shifts said B data three binary positions;

said transform means right shifts said B data one binary position;

said transform means adds said left shifted B data to said right shifted B data to obtain an intermediate value λ

4; and

said transform means adds said Y value to a negative λ

4 to obtain a value V.

50. A storage system for storing an image over a data channel, as in claim 48, wherein:

said transform means right shifts R data one binary position;

said transform means adds G data to said shifted R data to obtain an intermediate value λ

1;

said transform means left shifts said λ

1 value two binary positions;

said transform means adds said left shifted λ

1 value to said λ

1 value to obtain an intermediate value λ

2;

said transform means adds said λ

2 value to B data to obtain a value Y;

said transform means left shifts said R data three binary positions;

said transform means adds said left shifted R data to said right shifted R data to obtain an intermediate value λ

3;

said transform means adds said Y value to a negative λ

3 to obtain a value U;

said transform means left shifts said B data three binary positions;

said transform means right shifts said B data one binary position;

said transform means adds said left shifted B data to said right shifted B data to obtain an intermediate value λ

4; and

said transform means adds said Y value to a negative λ

4 to obtain a value V.