Detecting errors in video images

• US 5,745,169 A
• Filed: 01/18/1996
• Issued: 04/28/1998
• Est. Priority Date: 07/19/1993
• Status: Expired due to Term

• Current Assignee: British Telecommunications PLC (BT Group PLC)
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Murphy, Angela Ruth, Gunby, Stephen Richard
• Primary Examiner(s): Peng, John K.
• Assistant Examiner(s): Murrell, Jeffrey S.
• Application Number:
  US08/586,647
• Time in Patent Office: 831 Days
• Field of Search: 371/31, 348/403, 348/407, 348/421, 348/616, 348/404, 348/406, 348/408, 348/401, 348/402, 348/420, 348/192
• US Class Current: 348/192
• CPC Class Codes: H04N 19/60 using transform coding
• CPC Class Codes: H04N 19/65 using error resilience
• CPC Class Codes: H04N 19/89 involving methods or arrangements for detection of transmission errors at the decoder
• CPC Class Codes: H04N 19/895 in combination with error concealment

Abstract

An error is identified in a video image by calculating frequency-related coefficients for blocks of the image and considering the statistical distribution of said coefficients. The variance of the coefficients is calculated and a block of original image data is concealed if the variance calculated is significantly higher than a stored threshold and/or if said variance calculated is significantly different from variance values derived for similarly positioned blocks within a previous frame. Concealment is effected by selecting image data from a previous frame, displaced in response to transmitted motion vectors.

Claims

1. A method of detecting an error in a signal representing a video image, said method comprising:

receiving blocks of frequency related coefficients representing respective blocks of the image; and

analyzing the statistical distribution of said coefficients within a block to detect the presence of corrupted data.

2. A method as in claim 1, wherein said analysing step includes calculating the variance of said coefficients in a block.

3. A method as in claim 2, further comprising:

comparing the calculated variance of a block with a predetermined threshold value, an error being detected if the calculated variance exceeds the predetermined threshold value.

4. A method as in claim 2 further comprising:

comparing the calculated variance of a block of a current frame with the variance value(s) calculated for at least one similarly positioned block(s) in a previous frame, an error being detected if the calculated variance for a block of the current frame differs from the variance of the similarly positioned block(s) of a previous frame by a predetermined amount.

5. A method as in claim 1 further comprising:

firstly dividing the signal into a plurality of portions corresponding to respective blocks of an image; and

transforming said portions into blocks of frequency related coefficients.

6. Apparatus for identifying the presence of an error in a signal representing a video image, said apparatus comprising:

means for receiving blocks of frequency related coefficients representing respective blocks of the image; and

means for analysing the statistical distribution of said frequency related coefficients representing respective blocks of the image to identify corrupted data.

7. Apparatus as in claim 6, wherein said analysing means includes:

means for calculating the variance of the coefficients in a block.

8. Apparatus as in claim 7, wherein said analysing means includes:

means for comparing said variance value against a predetermined threshold value.

9. Apparatus as in claim 7, wherein said analysing means includes:

means for comparing said variance value against variance value(s) obtained for at least one similarly positioned block(s) in a previous frame.

10. Apparatus as in claim 6, including:

means for storing image data, and

means operable in response to detection of an error within a block to substitute for that block a stored block from an earlier frame of the image.

11. Apparatus as in claim 10, including:

means for reading motion vectors included with transmitted video data, and

processing means for concealing blocks identified as containing errors by selecting a stored block in response to said motion vectors.

12. Apparatus as in claim 11, wherein:

a new motion vector is calculated by averaging the motion vectors associated with similarly positioned blocks within the previous frame.

13. Apparatus as in claim 6, further comprising:

means for firstly dividing a signal into a plurality of portions corresponding to respective blocks of an image; and

means for transforming said portions into blocks of frequency related coefficients.