High resolution color video image acquisition and compression system

• US 4,743,959 A
• Filed: 09/17/1986
• Issued: 05/10/1988
• Est. Priority Date: 09/17/1986
• Status: Expired due to Fees

• Current Assignee: Frederiksen & SHU Laboratories Incorporated
• Original Assignee: Frederiksen & SHU Laboratories Incorporated
• Inventors: Frederiksen, Jeffrey E.
• Primary Examiner(s): Britton, Howard W.
• Assistant Examiner(s): Kostak, Victor R.
• Application Number:
  US06/908,316
• Time in Patent Office: 601 Days
• Field of Search: 358/133, 358/11, 358/13, 358/12, 358/96, 358/140, 358/22, 358/21 R, 358/280, 358/283, 358/284, 382/51, 382/56, 382/52, 364/518, 364/521, 379/53
• US Class Current: 375/240.25
• CPC Class Codes: G06T 9/005 Statistical coding, e.g. Huffman, run length coding
• CPC Class Codes: H04N 1/415 in which the picture-elements are subdivided or grouped into fixed one-dimensional or two-dimensional blocks
• CPC Class Codes: H04N 1/646 Transmitting or storing colour television type signals, e.g. PAL, Lab; Their conversion into additive or subtractive colour signals or vice versa therefor H04N1/642, H04N1/644 take precedence
• CPC Class Codes: H04N 11/042 Codec means

Abstract

A digital color video image digitizing and communication process comprises the steps of dividing digital data representing the overall picture information for a video image into sets of data representing shape and color information for blocks of the image with the color information including the major colors for the corresponding block and the variation in shades or hues of the colors within the block, separating, according to a predefined encoding procedure, each of the data sets into a first group of data representing the major colors constituting the block and a second group of data comprising the remaining data in the data sets and representing the shades of the major colors within the block, storing the second group of data in a color table, processing the second group of data and transmitting the first group of data and the processed second group of data to a decoder capable of processing received data according to a predefined decoding procedure, processing the received first data group to display the encoded image in its major colors, and combining the first group of data with the selected data from the second data group and processing the combined data to display the encoded video image with the variations in shades of the major colors. The digitizing process is adapted to operate in a plurality of modes incorporating different encoding/decoding procedures and provides high resolution digitizing of color images while at the same time providing the capability of controlling the degree of compression at the encoding end as well as the decoding end, thereby making it suited to a variety of color image digitizing applications.

Claims

1. A color video image acquisition and compression process for encoding and decoding color images, said process comprising the steps of:

acquiring digital data representing luminance and chrominance components of a color video image,

dividing the digital data representing the color video image into data sets representing blocks of the color image, each block containing a plurality of pixels each of which has a luminance value Y and a plurality of chrominance values,

calculating the average luminance value Yav for all pixels in the block,

determining a representative luminance value Y1 within a range extending between the maximum and average luminance values of pixels having luminance values above the average value Yav in each block, and determining a representative luminance value Y0 within a range extending between the minimum and average luminance values of pixels having luminance values below the average luminance value Yav in each block,

generating picture shape information for each of said blocks by identifying and encoding a contour associated with said representative luminance values Y0 and Y1, said contour representing clusters of pixels having the same representative luminance value Y0 or Y1,

comparing said encoded contour for each block with a series of predetermined encoded contours and selecting the predetermined encoded contour which corresponds most closely to the actual contour encoded for the block,

transmitting to a decoder encoded data comprising said representative luminance values Y1 and Y0, corresponding chrominance values, and the selected predetermined encoded contour for each block, and

decoding the transmitted data to reproduce the original color image.

2. The process of claim 1 which includes the step of calculating at least one representative pair of chrominance values for each block.

3. The process of claim 1 wherein said chrominance values comprise at least one set of red, blue and green values for each block.

4. The process of claim 1 which includes the steps of

calculating a first measure of error resulting from the encoding of the luminance information for a block,

determining whether said first error exceeds a first predetermined threshold value, and

if said first error exceeds said first threshold value, transmitting to said decoder all said luminance values Y for that block to produce a higher resolution reproduction of that portion of the original image.

5. The process of claim 4 which includes the steps of

processing the luminance information for a block to produce a bit map capable of reproducing that portion of the original image with a higher resolution than said encoded contour but a lower resolution than all said luminance values for that block, said bit map corresponding to the array of pixels within a block and representing partial luminance information for the different pixels,

if said first measure of error does not exceed said first threshold value, calculating a second measure of error resulting from the encoding of the luminance information for a block,

determining whether said second error exceeds a second predetermined threshold value, and

if said second error exceeds said second threshold value, transmitting to said decoder said encoded luminance data.

6. A digital color video image encoding and decoding process comprising the steps of:

dividing digital data representing the overall color information for a video image into sets of data representing color information for blocks of the image,

separating each color-information data set into (i) a first sub-set of encoded data representing sufficient luminance and chrominance information to permit a mosaic reproduction of the original image, (the "mosaic data"), said mosaic data including a representative luminance value for an entire block, and (ii) a second sub-set of encoded data representing additional luminance information to permit the original image to be reproduced with a greater degree of color resolution than said mosaic reproduction (the "high-resolution data"),

said high-resolution data being stored in a color table, said mosaic data being used as the address of the color table location containing the corresponding high-resolution data,

transmitting and decoding said mosaic data to display a mosaic reproduction of the original image, and

periodically transmitting and decoding said high-resolution data to display a reproduction of the original image with a greater degree of color resolution than said mosaic reproduction.

7. The process of claim 6 wherein all the high-resolution data having a common address in the color table is averaged to compress the amount of data stored in the color table.

8. The process of claim 6 wherein said high-resolution data includes a bit map corresponding to the array of pixels within a block and representing partial luminance information for the different pixels.

9. A digital color video image encoding and decoding process comprising the steps of:

dividing digital data representing the overall color information for a video image into sets of data representing color information for blocks of the image,

separating each color-information data set into (i) a first sub-set of encoded data representing sufficient luminance and chrominance information to permit a mosaic reproduction of the original image, (the "mosaic data") said mosaic data including a representative luminance value for an entire block, said value representing the most significant chrominance information for the block, and (ii) a second sub-set of encoded data representing additional luminance informawtion to permit the original image to be reproduced with a greater degree of color resolution than said mosaic reproduction (the "high-resolution data") said high-resolution data including the least significant chrominance information for a block,

transmitting and decoding said mosaic data to display a mosaic reproduction of the original image, and

periodically transmitting and decoding said high-resolution data to display a reproduction of the original image with a greater degree of color resolution than said mosaic reproduction.

10. A digital color video image encoding and decoding process comprising the steps of:

dividing digital data representing the overall color information for a video image into sets of data representing color information for blocks of the image,

separating each color-information data set into (i) a first sub-set of encoded data representing sufficient luminance and chrominance information to permit a mosaic reproduction of the original image, (the "mosaic data"), said mosaic data including a representative luminance value for an entire block, and (ii) a second sub-set of encoded data representing additional luminance information to permit the original image to be reproduced with a greater degree of color resolution than said mosaic reproduction (the "high-resolution data"),

said high-resolution data including information for determining a representative luminance value Y0 within a range extending between the maximum and average luminance values of pixels having luminance values above the average value Yav in each block, and for determining a representative luminance value Y1 within a range extending between the minimum and average luminance values of pixels having luminance values below the average luminance value Yav in each block; and

information representing the contour of said Y0 and Y1 values within a block,

transmitting and decoding said mosaic data to display a mosaic reproduction of the original image, and

periodically transmitting and decoding said high-resolution data to display a reproduction of the original image with a greater degree of color resolution than said mosaic reproduction.

11. A color video image acquisition and compression process for encoding color images, said process comprising the steps of:

acquiring digital data representing luminance and chrominance components of a color video image,

dividing digital data representing a color video image into data sets representing blocks of the color image, each block containing a plurality of pixels each of which has a luminance value Y and a plurality of chrominance values,

calculating the average luminance value Yav for all pixels in the block,

determining a representative luminance value Y1 between the maximum and average luminance values of pixels having luminance values above the average value Yav in each block, and determining a representative luminance value Y0 between the minimum and average luminance values of pixels having luminance values below the average luminance value Yav in each block,

calculating a median luminance value Ym midway between said values Y0 and Y1 for each block, calculating the difference Δ

Y between Ym and Y0 or Y1, and

generating picture shape information for each block.

12. The process of claim 11 which includes the steps of

transmitting and decoding Ym and the chrominance information for each block to display a mosaic reproduction of the original image in each frame, and

transmitting and decoding the bit map, Δ

Y and any remaining portions of Ym and the chrominance information for each block to display a high-resolution reproduction of the original image in selected frames.

13. The process of claim 11 wherein said picture shape information comprises an encoded contour.

14. The process of claim 11 wherein said picture shape information comprises a bit map in which each bit corresponds to one of the pixels in the array of pixels within a block, the value of each bit representing either a Y0 or Y1 value for each pixel.

15. The process of claim 11 further comprising the steps of:

calculating a measure of deviation between said pixel luminance values and said representative luminance values Y1 or Y0 for each block,

comparing said measure of deviation with a predefined threshold value, and

if said measure of deviation exceeds said predefined threshold value, dividing each block into two or more smaller blocks.

16. The process of claim 11 which includes the steps of:

calculating a single pair of representative chrominance values for each Ym,

utilizing a selected number of the most significant bits of Ym and said chrominance values to represent a fractional color volume,

utilizing said fractional color volume representation as an address in a color table,

storing at said address in a color table the value of Δ

Y and the remaining least significant bits of Ym and said chrominance values for each of said blocks, said stored values representing the luminance and chrominance positions within said fractional color volume for said block,

transmitting to a decoder said fractional color volume representations and the values stored in corresponding locations of the color table,

using said transmitted fractional color volume representations table for each of said blocks as an index into said transmitted color table to retrieve the values stored therein,

combining, according to a predetermined relationship, each of said fractional color volume representations with the corresponding values retrieved from said color table, and

using said combined values to reproduce the luminance and chrominance components of the original image.

17. The process of claim 11 wherein said picture shape information includes a contour bit map corresponding to the pixels in each block, said contour bit map having a value of 1 for pixels having a luminance value above said average luminance value Yav and a value of 0 for pixels having a luminance value below said average luminance value Yav.

18. The process of claim 16 wherein said decoder is used to display a mosaic of the original image by using only said fractional color volume representations transmitted to said decoder.

19. A digital color video image encoding and decoding process comprising the steps of:

acquiring digital data representing luminance and chrominance components of a color video image,

dividing digital data representing the overall color information for a video image into sets of data representing color information for blocks of the image,

separating each color-information data set into (i) a first sub-set of encoded data representing sufficient luminance and chrominance information to permit a mosaic reproduction of the original image, (the "mosaic data"), said mosaic data including a representative luminance value for an entire block, and (ii) a second sub-set of encoded data representing additional luminance information to permit the original image to be reproduced with a greater degree of color resolution than said mosaic reproduction (the "high-resolution data"),

storing said high-resolution data in a color table memory having addresses represented by said mosaic data so that said mosaic data can be used to identify the color table location containing the corresponding high-resolution data.

transmitting and decoding said mosaic data to display a mosaic reproduction of the original image, and

periodically transmitting and decoding said high-resolution data to display a reproduction of the original image with a greater degree of color resolution than said mosaic reproduction.

20. The process of claim 19 wherein all the high-resolution data having a common address in the color table is averaged to compress the amount of data stored in the color table.

21. The process of claim 19 wherein said data stored in the color table for blocks of the image being encoded is compared to the corresponding data stored in the color table for the preceding encoded image, and only the data for blocks which have new values is selected for transmission to the decoder.

22. A color video image acquisition and compression process for encoding and decoding color images, said process comprising the steps of:

acquiring digital data representing luminance and chrominance components of a color video image,

dividing the digital data representing the color video image into data sets representing blocks of the color image, each block containing a plurality of pixels each of which has a luminance value Y and a plurality of chrominance values,

calculating the average luminance value Yav for all pixels in the block,

determining a representative luminance value Y1 within a range extending between the maximum and average luminance values of pixels having luminance values above the average value Yav in each block, and determining a representative luminance value Y0 within a range extending between the minimum and average luminance values of pixels having luminance values below the average luminance value Yav in each block,

generating picture shape information for each of said blocks,

calculating a first measure of error between the original luminance values of the pixels in a block and said representative luminance values Y0 and Y1,

determining whether said first error exceeds a first predetermined threshold value,

transmitting to a decoder encoded data comprising said representative luminance values Y1 and Y0, corresponding chrominance values, and the picture shape information for each block, and if said first error exceeds said first threshold value, transmitting to said decoder all said luminance values Y for that block to produce a higher resolution reproduction of that portion of the original image, and

decoding the transmitted data to reproduce the original color image.

23. The process of claim 22 wherein said picture shape information is generated by identifying and encoding a contour for each block, and which includes the additional steps of

processing the luminance information for a block to produce encoded luminance data capable of reproducing that portion of the original image with a higher resolution than said encoded contour but a lower resolution than all said luminance values for that block,

if said first measure of error does not exceed said first threshold value, calculating a second measure of error resulting from the encoding of the luminance information for a block,

determining whether said second error exceeds a second predetermined threshold value, and

if said second error exceeds said second threshold value, transmitting to said decoder said encoded luminance data.

24. The process of claim 22 which includes the step of calculating at least one representative pair of chrominance values for each block.

25. The process of claim 22 wherein said chrominance values comprise at least one set of red, blue and green values for each block.

26. A color video image acquisition and compression process for encoding and decoding color images, said process comprising the steps of:

acquiring digital data representing luminance and chrominance components of a color video image,

dividing the digital data representing the color video image into data sets representing sectors of the color image, each sector containing a plurality of blocks of pixels each of which has a luminance value Y and a plurality of chrominance values,

calculating the average luminance value Yav for all pixels in the sector,

determining a representative luminance value Y1 within a range extending between the maximum and average luminance values of pixels having luminance values above the average value Yav in each sector, and determining a representative luminance value Y0 within a range extending between the minimum and average luminance values of pixels having luminance values below the average luminance value Yav in each sector,

generating representative chrominance values for each of said sectors,

generating picture shape information for each of said sectors,

calculating a measure of error between the color values of the pixels in a sector and said representative luminance or chrominance values,

determining whether said error exceeds a predetermined threshold value,

if said error is below said threshold value, transmitting to a decoder encoded data comprising luminance values, chrominance values, and picture shape information for that sector,

if said error exceeds said first threshold value, dividing the digital data representing the color video image into data sets representing blocks of the color image, each block containing a plurality of pixels each of which has a luminance value Y and a plurality of chrominance values,

calculating the average luminance value Yav for all pixels in the block,

determining a representative luminance value Y1 within a range extending between the maximum and average luminance values of pixels having luminance values above the average value Yav in each block, and determining a representative luminance value Y0 within a range extending between the minimum and average luminance values of pixels having luminance values below the average luminance value Yav in each block,

generating representative chrominance values for each of said blocks,

generating picture shape information for each of said blocks,

transmitting to said decoder the block-encoded data comprising luminance values, chrominance values, and picture shape information for that block, and

decoding the transmitted data to reproduce the original color image.

27. The process of claim 26 wherein said measure of error is representative of the error between the original luminance values of the pixels and said representative luminance values Y0 and Y1 for the sector.

28. The process of claim 26 wherein said measure of error is representative of the error between the original chrominance values of the pixels and said representative chrominance values of the sectors.