Apparatus and method for adaptively compressing successive blocks of digital video

• US 5,091,782 A
• Filed: 04/09/1990
• Issued: 02/25/1992
• Est. Priority Date: 04/09/1990
• Status: Expired due to Term

• Current Assignee: CIF Licensing LLC
• Original Assignee: General Instrument Corporation (CommScope Holding Company, Inc.)
• Inventors: Krause, Edward A., Paik, Woo H.
• Primary Examiner(s): Peng, John K.
• Application Number:
  US07/507,258
• Time in Patent Office: 687 Days
• Field of Search: 358/133, 358/135, 358/136, 358/105, 382/56
• US Class Current: 348/400.1
• CPC Class Codes: H04N 19/112 according to a given display mode, e.g. for interlaced or progressive display mode
• CPC Class Codes: H04N 19/136 Incoming video signal characteristics or properties
• CPC Class Codes: H04N 19/154 Measured or subjectively estimated visual quality after decoding, e.g. measurement of distortion use of rate-distortion criteria H04N19/147
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/60 using transform coding
• CPC Class Codes: H04N 19/61 in combination with predictive coding

Abstract

The compression of successive blocks of digital data is optimized by selecting between different compression algorithms or different data formats on a block-by-block basis. In one application, digitized interlaced video signals are processed for transmission in a compressed form. A set of pixel data presented in a field format is compressed to provide a first compressed video signal. The set of pixel data is also presented in a frame format and compressed to provide a second compressed video signal. Errors are evaluated in the first and second compressed video signals. The compressed video signal having the least error is selected for further processing. The technique is repeated for successive sets of pixel data and the selected signals are encoded to identify them as field formatted or frame formatted signals. The encoded selected signals are then combined to provide a compressed video signal data stream for transmission. Apparatus for receiving and decoding the signals is also disclosed.

Claims

1. Apparatus for processing digitized interlaced video signals for transmission in a compressed form comprising:

first means for compressing a set of pixel data presented in a field format to provide a first compressed video signal;

second means for compressing said set of pixel data presented in a frame format to provide a second compressed video signal;

means coupled to said first means for evaluating errors in the first compressed video signal and coupled to said second means for evaluating errors in the second compressed video signal; and

means responsive to said error evaluating means for selecting the compressed video signal having the least error.

2. Apparatus in accordance with claim 1 wherein:

successive sets of pixel data are sequentially compressed and evaluated; and

said selecting means selects the compressed video signal having the least error for each particular set.

3. Apparatus in accordance with claim 2 further comprising:

means for encoding the selected signals to identify them as field formatted or frame formatted signals.

4. Apparatus in accordance with claim 3 further comprising:

means for combining the encoded selected signals to provide a compressed video signal data stream for transmission.

5. Apparatus in accordance with claim 4 further comprising:

means for dividing a digitized interlaced video signal into even and odd blocks of pixel data corresponding to even and odd fields of a video frame; and

means for presenting said even and odd blocks to said first compressing means in an alternating order defining said field format.

6. Apparatus in accordance with claim 5 further comprising:

means for grouping said blocks into corresponding odd/even block pairs and scanning the odd and even lines of each pair in an alternating order to provide interleaved lines of pixel data; and

means for presenting the interleaved lines of pixel data from successive block pairs to said second compressing means.

7. Apparatus in accordance with claim 6 wherein said sets each comprise two vertically adjacent blocks of pixel data from a video frame.

8. Apparatus in accordance with claim 1 wherein:

said first and second compressing means produce and quantize respective first and second arrays of transform coefficients for the set of pixel data; and

said error evaluating means determines the error between the quantized transform coefficients and the unquantized transform coefficients of each array.

9. Apparatus in accordance with claim 8 wherein:

said error evaluating means determines said error by computing the difference between each coefficient in an array before and after quantization; and

said selecting means compares the average error of all the coefficient differences in said first array to the average error of all the coefficient differences in said second array to identify the array having the least total error.

10. Apparatus in accordance with claim 1 wherein:

said first compressing means produces a first array of transform coefficients for said set of pixel data;

said second compressing means produces a second array of transform coefficients for said set of pixel data; and

said error evaluating means comprises;

means for inverse transforming said first and second arrays to recover the pixel data, and

means for determining the error between the pixel data recovered from each array and the original set of pixel data presented to the first and second compressing means.

11. Apparatus in accordance with claim 1 further comprising:

means for receiving a digitized interlaced video signal containing pixel data defining a sequence of video frames;

means for predicting the pixel data for a current video frame from pixel data of a previous video frame;

means for subtracting the predicted pixel data from the actual pixel data for the current video frame to produce a set of pixel data representing a prediction error;

means for presenting the prediction error pixel data to said first compressing means in a field format; and

means for presenting the prediction error pixel data to said second compressing means in a frame format.

12. Apparatus in accordance with claim 11 further comprising:

means for dividing each video frame of the digitized interlaced video signal into successive blocks of pixel data for processing on a block-by-block basis by said predicting, subtracting and presenting means.

13. Apparatus in accordance with claim 11 further comprising:

means for encoding the selected signals with motion vector data generated by said predicting means.

14. Apparatus in accordance with claim 13 further comprising:

means for encoding the selected signals to identify them as field formatted or frame formatted signals.

15. Apparatus in accordance with claim 14 further comprising:

means for combining the encoded selected signals to provide a compressed video signal data stream for transmission.

16. Apparatus for optimizing the compression of successive blocks of digital data comprising:

means for compressing data blocks in a first data path to provide a first compressed signal;

means for compressing said data blocks in a second data path to provide a second compressed signal;

means for evaluating errors in the first and second compressed signals;

means responsive to said evaluating means for selecting the compressed signal having the least error for each data block; and

means for combining the selected signals to provide a compressed digital data stream.

17. Apparatus in accordance with claim 16 further comprising:

means for encoding each selected signal with data indicative of the data path in which the signal originated.

18. A receiver for decoding the compressed digital data stream from the apparatus of claim 17 comprising:

means for detecting the encoding data from each selected signal to identify whether the signal originated in the first data path or the second data path; and

means responsive to said detecting means for processing the selected signals from the first data path in a corresponding first decompression path and for processing the selected signals from the second data path in a corresponding second decompression path.

19. Apparatus in accordance with claim 18 wherein:

said first data path and first decompression path use a first algorithm for compression and decompression of said data blocks; and

the second data path and second decompression path use a second algorithm for compression and decompression of said data blocks.

20. Apparatus in accordance with claim 16 wherein:

said data blocks are provided to said first data path in a first format and to said second data path in a second format.

21. Decoder apparatus comprising:

means for receiving compressed digital video signals transmitted in blocks of frame processed pixel data and field processed pixel data;

means coupled to said receiving means for determining whether a particular block of data contained in a received signal was frame processed or field processed;

first means for decoding received blocks of field processed pixel data;

second means for decoding received blocks of frame processed pixel data; and

means responsive to said determining means for selectively combining decoded blocks from said first and second means to recover an uncompressed video signal.

22. Apparatus in accordance with claim 21 further comprising means for converting the recovered signal to a digitized interlaced video signal.

23. Apparatus in accordance with claim 21 wherein said determining means comprises means for reading field and frame processing identification data appended to said blocks.

24. Apparatus in accordance with claim 21 wherein said blocks comprise arrays of transform coefficients, and the first and second means comprise means for inverse transforming said coefficients.

25. Apparatus in accordance with claim 24 wherein the inverse transforming means is shared by said first and second means.

26. Apparatus in accordance with claim 21 further comprising:

means for retrieving motion vector data appended to received blocks of pixel data representing a current video frame;

means for storing data representing a previous video frame;

means for computing prediction signals from the retrieved motion vector data and the stored data; and

means for adding said prediction signals to the received blocks for the current video frame.

27. A digital television system comprising:

first data compression means for compressing field formatted blocks of pixel data derived from a digitized interlaced video signal;

second data compression means for compressing frame formatted blocks of pixel data corresponding to said field formatted blocks;

means for evaluating errors in compressed data from said first and second data compression means;

means responsive to said error evaluating means for selecting the compressed data for each block having the least error;

means for encoding the selected data for each block to identify it as field processed or frame processed data; and

means for combining the encoded selected data to provide a compressed video data stream containing interspersed blocks of field processed pixel data and frame processed pixel data for transmission by a transmitter.

28. The system of claim 27 further comprising:

receiver means for receiving said compressed video data stream from said transmitter;

means operatively associated with said receiver means for decoding the encoded selected data in said data stream to identify field processed blocks and frame processed blocks;

first means for processing received blocks of field processed data;

second means for processing received blocks of frame processed data; and

means responsive to said decoding means for selectively combining blocks from said first and second processing means to recover an uncompressed digitized interlaced video signal.

29. The system of claim 28 wherein said digitized interlaced video signal defines a sequence of video frames, said system further comprising:

means for predicting the pixel data for a current video frame from pixel data of a previous video frame;

means for subtracting the predicted pixel data from the actual pixel data for the current video frame to produce a set of pixel data representing a prediction error;

means for presenting the prediction error pixel data to said first data compression means as field formatted blocks;

means for presenting the prediction error pixel data to said second data compression means as frame formatted blocks;

means operatively associated with said predicting means for generating motion vector data for the current video frame;

means operatively associated with said generating means for encoding the selected data for each block with corresponding motion vector data;

means operatively associated with said receiver means for retrieving the motion vector data from each block of a current video frame;

means operatively associated with said receiver for storing data representing a previous video frame;

means for computing prediction signals from the retrieved motion vector data and the stored data; and

means for adding said prediction signals to the received blocks for the current video frame.

30. A method for encoding television signals for digital transmission comprising the steps of:

dividing a digitized interlaced video signal into blocks of pixel data;

formatting said blocks in a field format;

compressing said field formatted blocks to provide a first compressed video signal;

formatting said blocks in a frame format;

compressing said frame formatted blocks to provide a second compressed video signal;

evaluating errors in said first and second compressed video signals;

selecting, for each block, the compressed video signal having the least error;

encoding the selected signals to identify them as field formatted or frame formatted signals; and

combining the encoded signals.

31. A method in accordance with claim 30 comprising the further steps of:

predicting the pixel data for a current video frame contained in said digitized interlaced signal from pixel data of a previous frame;

subtracting said predicted pixel data from the actual pixel data for the current frame to provide an abbreviated set of pixel data for use in producing said first and second compressed video signals; and

encoding the selected signals with motion vector data generated during said prediction step.

32. A method for decoding the combined encoded signals produced by the method of claim 31, comprising the steps of:

decompressing field formatted signals in a decompression path adapted for field processing of data;

decompressing frame formatted signals in a decompression path adapted for frame processing of data;

retrieving the motion vector data from encoded selected signals representing a current video frame;

storing data representing a previous video frame;

computing prediction signals from the retrieved motion vector data and the stored data;

adding said prediction signals to the decompressed signals; and

combining the decompressed signals to recover the digitized interlaced video signal.

33. A method for decoding the combined encoded signals produced by the method of claim 30, comprising the steps of:

decompressing field formatted signals in a decompression path adapted for field processing of data;

decompressing frame formatted signals in a decompression path adapted for frame processing of data; and

combining the decompressed signals to recover the digitized interlaced video signal.