Method of decoding coded image data utilizing transmission error detection to interrupt decoding and up-dating

• US 5,117,288 A
• Filed: 11/27/1990
• Issued: 05/26/1992
• Est. Priority Date: 11/27/1989
• Status: Expired due to Fees

• Current Assignee: Licentia Patent-Verwaltungs-GmbH (AB Electrolux)
• Original Assignee: Licentia Patent-Verwaltungs-GmbH (AB Electrolux)
• Inventors: Wolf, Winfried, May, Franz, Kummerfeldt, Georg, Eisenhardt, Volker
• Primary Examiner(s): Kostak, Victor R.
• Application Number:
  US07/618,566
• Time in Patent Office: 546 Days
• Field of Search: 358/135, 358/136, 358/133, 371/55, 371/30, 371/31, 371/32, 371/34
• US Class Current: 375/240.12
• CPC Class Codes: H04N 19/50 using predictive coding H04N19/61 takes precedence
• CPC Class Codes: H04N 19/60 using transform coding
• CPC Class Codes: H04N 19/65 using error resilience
• CPC Class Codes: H04N 19/895 in combination with error concealment

Abstract

A method of decoding coded image data, preferably coded according to CCITT Recommendsation H. 261, wherein the data stream is continuously monitored for transmission errors by utilizing the residual redundancy of the data stream, and pixel values reconstructed after the detection of a transmission error are not entered into the predicted-image memory until a subsequent starting code word for a frame or field has appeared.

Claims

1. In a method for decoding and utilizing coded image data from a received data stream, particularly for the transmission of sequences of moving images, wherein the image date are coded according to a coding method employing residual redundancy and the data stream includes unequivocally associatable starting code words for a frame or fields, said method including the steps of decoding the image data to provide reconstructed pixel values, and up-dating the contents of a predicted-image memory on the basis of the reconstructed pixel values obtained form said step of decoding;

the improvement comprising;

following detection of one of said starting code words, constantly checking said data stream, utilizing said residual redundancy of said image data, for the occurrence of a transmission error; and

, upon the detection of a transmission error, interrupting said step of up-dating, and no longer utilizing said reconstructed pixel values to bring the predicted-image memory up to date, until a new starting code word appears for the frame or field.

2. A method as defined in claim 1, wherein the residual redundancy of the coding method is such that not all bit sequences are permissible;

and said step of constantly checking includes carrying out a continuous check during said step of decoding to determine whether a decoded signal has left permissible paths of a decoding tree, and detecting a transmission error if the decoded signal has left permissible paths of the decoding tree.

3. A method as defined in claim 2, further comprising utilizing error reports from other components of a receiver to additionally indicate the occurrence of a transmission error and to stop the utilization of reconstructed pixel values for bringing the predicted-image memory content up to date until the occurrence of a new starting code word for the frame or field.

4. A method as defined in claim 3, further comprising:

after detection of a transmission error, excluding pixel values reconstructed before said detection of said error for a given time period from being utilized for up-dating the predicted-image memory content.

5. A method as defined in claim 2, further comprising:

after detection of a transmission error, excluding pixel values reconstructed before said detection of said error for a given time period from being utilized for up-dating the predicted-image memory content.

6. A method as defined in claim 1, further comprising utilizing error reports from other components of a receiver to additionally indicate the occurrence of a transmission error and to stop the utilization of reconstructed pixel values for bringing the predicted-image memory content up to date until the occurrence of a new starting code word for the frame or field.

7. A method as defined in claim 6, further comprising:

after detection of a transmission error, excluding pixel values reconstructed before said detection of said error for a given time period from being utilized for up-dating the predicted-image memory content.

8. A method as defined in claim 7, wherein said image data are coded according to CCITT Recommendation H. 261 in which each frame is subdivided into a number of fields which are in turn subdivided into a plurality of macro-blocks which are divided into six blocks each containing 8×

8 pixels.

9. A method as defined in claim 8, wherein, after detection of a transmission error, all already reconstructed pixel values of the momentary macro block and also pixel values of the last preceding block or macro block are additionally not utilized to bring the predicted-image memory up to date.

10. A method as defined in claim 6, wherein a transmission error is determined to have occurred if a decoded and dequantized spectral component lies outside of a permissible amplitude range for spectral components.

11. A method as defined in claim 6, wherein each field is divided into a given number of macro-blocks and a transmission error is determined to have occurred if an unduly high macro block number appears.

12. A method as defined in claim 6, wherein a transmission error is determined to have occurred if shift vectors point to pixels outside of the coded image region.

13. A method as defined in claim 6, wherein each field is divided into a number of macro-blocks each of which in turn is subdivided into a plurality of blocks with a given number of pixels, and a transmission error is determined to have occurred if more than a given maximum number of spectral components are decoded in a block.

14. In a method for decoding and utilizing coded image data from a received data stream, particularly for the transmission of sequences of moving images, wherein the image data are coded according to a coding method employing residual redundancy and the data stream includes unequivocally associatable starting code words for a frame or fields, said method including the steps of decoding the image data to provide reconstructed pixel values, adding the reconstructed pixel values obtained from said step of decoding to the output of a predicted-image memory to provide a digital video signal, and up-dating the contents of the predicted image memory with the digital video signal;

the improvement comprising;

following detection of one of said starting code words, continuously checking said data stream, utilizing said residual redundancy of said image data, for the occurrence of a transmission error; and

, upon the detection of a said transmission error, immediately interrupting said steps of adding and up-dating until a new starting code word is detected for a frame or field.

15. A method as defined in claim 14, wherein the residual redundancy of the coding method is such that not all bit sequences are permissible;

and said step of checking includes carrying out a continuous check during said step of decoding to determine whether a decoded signal has left permissible paths of a decoding tree, and detecting a transmission error if the decoded signal has left permissible paths of the decoding tree.

16. A method as defined in claim 14, wherein said occurrence of a transmission error is detected if a decoded and dequantized spectral component has a value outside of a permissible amplitude value range for spectral components.

17. A method as defined in claim 14, wherein said occurrence of a transmission error is detected if decoded shift vectors point to pixels outside of the coded image region being evaluated.

18. A method as defined in claim 14, wherein each field is divided into a number of blocks each having a given number of pixels, and said occurrence of a transmission error is detected if the number of spectral components decoded in a block is greater than said given number of pixels.