Methods and systems for estimation of compression noise

• US 8,588,535 B2
• Filed: 09/15/2010
• Issued: 11/19/2013
• Est. Priority Date: 09/15/2010
• Status: Expired due to Fees

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: Sharp Laboratories of America Incorporated (Hon Hai Precision Industry Co., Ltd.)
• Inventors: Segall, Christopher A.
• Primary Examiner(s): Krasnic, Bernard
• Application Number:
  US12/882,763
• Publication Number:
  US 20120063694A1
• Time in Patent Office: 1,161 Days
• Field of Search: 382/232
• US Class Current: 382/232
• CPC Class Codes: G06T 2207/20192 Edge enhancement; Edge preservation
• CPC Class Codes: G06T 2207/30168 Image quality inspection
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: H04N 19/86 involving reduction of coding artifacts, e.g. of blockiness

Abstract

Systems and methods for determining the compression noise present in an image or frame in a video sequence are provided. Horizontal and vertical pixel value differences are computed. A plurality of horizontal standard deviation values associated with a plurality of horizontal offsets are computed, and a plurality of vertical standard deviation values associated with a plurality of vertical offsets are computed. Two horizontal statistics are computed using the plurality of horizontal standard deviation values, and two vertical statistics are computed using the plurality of vertical standard deviation values. The two horizontal statistics are combined to form an estimate of horizontal compression noise, and the two vertical statistics are combined to form an estimate of vertical compression noise. The estimate of horizontal compression noise is combined with the estimate of vertical compression noise to form a compression noise estimate.

Claims

1. A method for estimating compression noise in an image, said method comprising:

a) receiving an image in a noise-estimation system;

b) computing a horizontal difference value for each pixel in a first plurality of pixels in said image, thereby producing a plurality of horizontal difference values;

c) computing a vertical difference value for each pixel in a second plurality of pixels in said image, thereby producing a plurality of vertical difference values;

d) computing a plurality of horizontal standard deviation values, wherein each of said horizontal standard deviation values in said plurality of horizontal standard deviation values is associated with a horizontal offset within said image;

e) computing a plurality of vertical standard deviation values, wherein each of said vertical standard deviation values in said plurality of vertical standard deviation values is associated with a vertical offset within said image;

f) computing a first horizontal statistic using said plurality of horizontal standard deviation values;

g) computing a second horizontal statistic using said plurality of horizontal standard deviation values;

h) computing a first vertical statistic using said plurality of vertical standard deviation values;

i) computing a second vertical statistic using said plurality of vertical standard deviation values;

j) combining said first horizontal statistic and said second horizontal statistic to form an estimate of horizontal compression noise;

k) combining said first vertical statistic and said second vertical statistic to form an estimate of vertical compression noise; and

l) combining said estimate of horizontal compression noise and said estimate of vertical compression noise to form a compression noise estimate.

2. The method as described in claim 1, wherein said horizontal offset and said vertical offset are related to a coding block offset associated with a codec associated with said image.

3. The method as described in claim 1, wherein:

said first horizontal statistic is a mean of said horizontal standard deviation values; and

said first vertical statistic is a mean of said vertical standard deviation values.

4. The method as described in claim 1, wherein:

said second horizontal statistic is a maximum of said horizontal standard deviation values; and

said second vertical statistic is a maximum of said vertical standard deviation values.

5. The method as described in claim 4, wherein:

said first horizontal statistic is a mean of said horizontal standard deviation values; and

said first vertical statistic is a mean of said vertical standard deviation values.

6. The method as described in claim 1, wherein said combining said first horizontal statistic and said second horizontal statistic comprises computing a weighted average of said first horizontal statistic and said second horizontal statistic.

7. The method as described in claim 1, wherein said combining said first vertical statistic and said second vertical statistic comprises computing a weighted average of said first vertical statistic and said second vertical statistic.

8. The method as described in claim 1, wherein said combining said estimate of horizontal compression noise and said estimate of vertical compression noise comprises determining the maximum value of said estimate of horizontal compression noise and said estimate of vertical compression noise.

9. The method as described in claim 1, wherein said image is a first frame in a video sequence and further comprising:

estimating a horizontal block-boundary location associated with each frame in a plurality of frames in said video sequence, wherein said first frame is in said plurality of frames, thereby producing a plurality of horizontal block-boundary locations;

estimating a vertical block-boundary location associated with each frame in said plurality of frames in said video sequence, thereby producing a plurality of vertical block-boundary locations;

determining a number of unique horizontal block-boundary locations in said plurality of horizontal block-boundary locations;

determining a number of unique vertical block-boundary locations in said plurality of vertical block-boundary locations;

comparing said number of unique horizontal block-boundary locations in said plurality of horizontal block-boundary locations to a horizontal threshold;

comparing said number of unique vertical block-boundary locations in said plurality of vertical block-boundary locations to a vertical threshold;

setting said estimate of horizontal compression noise to zero when said number of unique horizontal block-boundary locations in said plurality of horizontal block-boundary locations and said horizontal threshold meet a horizontal criterion; and

setting said estimate of vertical compression noise to zero when said number of unique vertical block-boundary locations in said plurality of vertical block-boundary locations and said vertical threshold meet a vertical criterion.

10. The method as described in claim 1, wherein said receiving said image comprises:

receiving a bitstream associated with said image; and

decoding said bitstream.

11. The method as described in claim 1 further comprising making said compression noise estimate available to an edge-preserving filter, wherein a parameter value associated with said edge-preserving filter is based on said compression noise estimate.

12. The method as described in claim 11, wherein:

said edge-preserving filter is a bi-lateral filter; and

said parameter value is a value for a range sigma parameter associated with said bi-lateral filter.

13. The method as described in claim 1 further comprising combining said compression noise estimate with a second noise estimate associated with said image.

14. The method as described in claim 13, wherein said combining said compression noise estimate with said second noise estimate comprises determining the maximum value of said compression noise estimate and said second noise estimate.

15. The method as described in claim 13 further comprising making said combined noise estimate available to an edge-preserving filter, wherein a parameter value associated with said edge-preserving filter is based on said combined noise estimate.

16. The method as described in claim 15, wherein:

said edge-preserving filter is a bi-lateral filter; and

said parameter value is a value for a range sigma parameter associated with said bi-lateral filter.

17. The method as described in claim 15, wherein said parameter value associated with said edge-preserving filter is proportional to said combined noise estimate.

18. The method as described in claim 1 further comprising compensating said compression noise estimate based on a level of sharpening associated with an image enhancement system.

19. The method as described in claim 18 further comprising making said compensated compression noise estimate available to an edge-preserving filter, wherein a parameter value associated with said edge-preserving filter is based on said compensated compression noise estimate.

20. The method as described in claim 19, wherein said edge-preserving filter is a bi-lateral filter.

21. The method as described in claim 18 further comprising temporally filtering said compensated compression noise estimate.

22. The method as described in claim 21 further comprising making said temporally filtered compensated compression noise estimate available to an edge-preserving filter, wherein a parameter value associated with said edge-preserving filter is based on said temporally filtered compensated compression noise estimate.

23. The method as described in claim 22, wherein:

said edge-preserving filter is a bi-lateral filter; and

said parameter value is a value for a range sigma parameter associated with said bi-lateral filter.

24. An apparatus for estimating compression noise in an image, said apparatus comprising:

a) an image receiver for receiving an image in a noise-estimation system;

b) a horizontal-difference calculator for computing a horizontal difference value for each pixel in a first plurality of pixels in said image, thereby producing a plurality of horizontal difference values;

c) a vertical-difference calculator for computing a vertical difference value for each pixel in a second plurality of pixels in said image, thereby producing a plurality of vertical difference values;

d) a horizontal-standard-deviation calculator for computing a plurality of horizontal standard deviation values, wherein each of said horizontal standard deviation values in said plurality of horizontal standard deviation values is associated with a horizontal offset within said image;

e) a vertical-standard-deviation calculator for computing a plurality of vertical standard deviation values, wherein each of said vertical standard deviation values in said plurality of vertical standard deviation values is associated with a vertical offset within said image;

f) a first horizontal-statistic calculator for computing a first horizontal statistic using said plurality of horizontal standard deviation values;

g) a second horizontal-statistic calculator for computing a second horizontal statistic using said plurality of horizontal standard deviation values;

h) a first vertical-statistic calculator for computing a first vertical statistic using said plurality of vertical standard deviation values;

i) a second vertical-statistic calculator for computing a second vertical statistic using said plurality of vertical standard deviation values;

j) a horizontal-statistics combiner for combining said first horizontal statistic and said second horizontal statistic to form an estimate of horizontal compression noise;

k) a vertical-statistics combiner for combining said first vertical statistic and said second vertical statistic to form an estimate of vertical compression noise; and

l) a compression-noise combiner for combining said estimate of horizontal compression noise and said estimate of vertical compression noise to form a compression noise estimate.

25. The apparatus as described in claim 24, wherein said image is a first frame in a video sequence and further comprising:

a horizontal-block-boundary estimator for estimating a horizontal block-boundary location associated with each frame in a plurality of frames in said video sequence, wherein said first frame is in said plurality of frames, thereby producing a plurality of horizontal block-boundary locations;

a vertical-block-boundary estimator for estimating a vertical block-boundary location associated with each frame in said plurality of frames in said video sequence, thereby producing a plurality of vertical block-boundary locations;

a horizontal-block-boundary locator for determining a number of unique horizontal block-boundary locations in said plurality of horizontal block-boundary locations;

a vertical-block-boundary locator for determining a number of unique vertical block-boundary locations in said plurality of vertical block-boundary locations;

a first comparator for comparing said number of unique horizontal block-boundary locations in said plurality of horizontal block-boundary locations to a horizontal threshold;

a second comparator for comparing said number of unique vertical block-boundary locations in said plurality of vertical block-boundary locations to a vertical threshold;

a horizontal-noise-selector for setting said estimate of horizontal compression noise to zero when said number of unique horizontal block-boundary locations in said plurality of horizontal block-boundary locations and said horizontal threshold meet a horizontal criterion; and

a vertical-noise-selector for setting said estimate of vertical compression noise to zero when said number of unique vertical block-boundary locations in said plurality of vertical block-boundary locations and said vertical threshold meet a vertical criterion.