Method and apparatus for recovering quantized coefficients

• US 6,373,894 B1
• Filed: 02/17/1998
• Issued: 04/16/2002
• Est. Priority Date: 02/18/1997
• Status: Expired due to Term

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.), MediaTek, Inc.
• Original Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.), Sarnoff Corporation (SRI International, Inc.)
• Inventors: Sodagar, Iraj, Florencio, Dinei Afonso Ferreira
• Primary Examiner(s): Kelley, Chris
• Assistant Examiner(s): Vo, Tung T.
• Application Number:
  US09/024,084
• Time in Patent Office: 1,519 Days
• Field of Search: H04/B.166, 375/240.02, 375/240.03, 375/240.04, 375/240.18, 375/240.19, 375/240.24, 375/240.25, 375/240.26, 375/240.29, 382/232, 382/234, 382/236, 382/240, 382/248, 382/250, 382/268, 382/264, 382/166, 704/500, 704/503, 341/51, 341/87
• US Class Current: 375/240.19
• CPC Class Codes: H04N 19/10 using adaptive coding
• CPC Class Codes: H04N 19/102 characterised by the element, parameter or selection affected or controlled by the adaptive coding
• CPC Class Codes: H04N 19/122 Selection of transform size, e.g. 8x8 or 2x4x8 DCT; Selection of sub-band transforms of varying structure or type
• CPC Class Codes: H04N 19/124 Quantisation
• CPC Class Codes: H04N 19/137 Motion inside a coding unit, e.g. average field, frame or block difference
• CPC Class Codes: H04N 19/14 Coding unit complexity, e.g. amount of activity or edge presence estimation H04N19/146 takes precedence
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/18 the unit being a set of transform coefficients
• CPC Class Codes: H04N 19/63 using sub-band based transform, e.g. wavelets
• CPC Class Codes: H04N 19/85 using pre-processing or post-processing specially adapted for video compression

Abstract

An apparatus and a concomitant method is disclosed for recovering or adjusting quantized coefficients by using a nonlinear method. The method operates by fitting the received signal into one of several predefined classes, and adjusting the signal as appropriate to better fit the best suited class.

Claims

1. Method for adjusting decoded pixel values that are representative of an image, where said decoded pixel values have previously undergone a transform operation and a quantization operation, said method comprising the steps of:

(a) selecting a block of decoded pixel values;

(b) dividing said block into a plurality of subblocks;

(c) computing an adjustment for each of said subblocks;

(d) applying a forward transform to said plurality of adjustments to produce a plurality of transform coefficients;

(e) determining whether said plurality of adjustments should be applied to said decoded pixel values; and

(f) adjusting said decoded pixel values within each subblock in accordance with said plurality of adjustments, if said determining step (e) determines that said plurality of adjustments should be applied.

2. The method of claim 1, wherein said applying step (d) applies a 2-d Haar transform.

3. The method of claim 1, wherein said dividing step (b) divides said block into four subblocks.

4. The method of claim 1, wherein said selecting step (a) selects a block size that correlates to a block size employed by the transform operation.

5. The method of claim 1, wherein said determining step (e) comprises the step of:

(e1) comparing said plurality of transform coefficients with a quantization step employed by the quantization operation;

(e2)determining that said plurality of adjustments should be applied to said decoded pixel values within each of said subblocks, if all of said adjustments of all of said subblocks are below said quantization step; and

(e3) removing a subblock that corresponds to a highest adjustment, if at least one of said adjustments of said subblocks is above said quantization step, and repeating said steps (d) to (e3) for said remaining subblocks.

6. The method of claim 5, wherein said determining step (e) stops if said remaining subblocks fall below a threshold of remaining subblocks.

7. The method of claim 6, wherein said threshold of remaining subblocks is set at three.

8. The method of claim 1, wherein said adjustments in said computing step (c) is computed by minimizing a fitting function.

9. The method of claim 8, wherein said fitting function is a sum of absolute differences between a plurality of medians of said block and a target median.

10. The method of claim 1, wherein steps (a) through (f) are repeated for a next hierarchical level of the image.

11. Apparatus for adjusting decoded pixel values that are representative of an image, where said decoded pixel values have previously undergone a transform operation and a quantization operation, said apparatus comprising:

a first selecting means for selecting at least one block of decoded pixel values;

a first dividing means for dividing said at least one block into a plurality of subblocks;

computing means for computing an adjustment for each of said subblocks;

applying means for applying a forward transform to said plurality of adjustments to produce a plurality of transform coefficients;

a first determining means for determining whether said plurality of adjustments should be applied to said decoded pixel values; and

adjusting means for adjusting said decoded pixel values within each subblock in accordance with said plurality of adjustments, if said first determining means determines that said plurality of adjustments should be applied.

12. The apparatus of claim 11, wherein said forward transform of said applying means is a 2-d Haar transform.

13. The apparatus of claim 11, wherein said first determining means comprises:

means for comparing said plurality of transform coefficients with a quantization step employed by the quantization operation;

a second means for determining that said plurality of adjustments should be applied to said decoded pixel values within each of said subblocks, if all of said adjustments of all of said subblocks are below said quantization step; and

mean for removing a subblock that corresponds to a highest adjustment, if at least one of said adjustments of said subblocks is above said quantization step.

14. The apparatus of claim 11, wherein said adjustments in said computing means is computed by minimizing a fitting function.

15. The apparatus of claim 14, wherein said fitting function is a sum of absolute differences between a plurality of medians of said block and a target median.

16. A decoding system for adjusting decoded pixel values that are representative of an image, where said decoded pixel values have previously undergone a transform operation and a quantization operation, said decoding system comprising:

an inverse quantizer for applying an inverse quantization process to the plurality of coefficients to produce a plurality of unquantized coefficients;

an inverse transform for applying an inverse transform operation to said plurality of dequantized coefficients to produce the decoded pixel values representative of the image; and

a coefficient recovery section for adjusting said decoded pixel values representative of the image, wherein said coefficient recovery section comprises;

a first selecting means for selecting at least one block of decoded pixel values;

a first dividing means for dividing said at least one block into a plurality of subblocks;

computing means for computing an adjustment for each of said subblocks;

applying means for applying a forward transform to said plurality of adjustments to produce a plurality of transform coefficients;

a first determining means for determining whether said plurality of adjustments should be applied to said decoded pixel values; and

adjusting means for adjusting said decoded pixel values within each subblock in accordance with said plurality of adjustments, if said first determining means determines that said plurality of adjustments should be applied.

17. The decoding system of claim 16, wherein said coefficient recovery section comprises:

a selecting means for selecting at least one block of decoded pixel values;

a dividing means for dividing said at least one block into a plurality of subblocks;

means for computing an adjustment for each of said subblocks;

means for applying a forward transform to said plurality of adjustments to produce a plurality of transform coefficients;

a first means for determining whether said plurality of adjustments should be applied to said decoded pixel values; and

means for adjusting said decoded pixel values within each subblock in accordance with said plurality of adjustments, if said first determining means determines that said plurality of adjustments should be applied.

18. The system of claim 17, wherein said forward transform of said applying means is a 2-d Haar transform.

19. The system of claim 17, wherein said first determining means comprises:

means for comparing said plurality of transform coefficients with a quantization step employed by the quantization operation;

a second means for determining that said plurality of adjustments should be applied to said decoded pixel values within each of said subblocks, if all of said adjustments of all of said subblocks are below said quantization step; and

mean for removing a subblock that corresponds to a highest adjustment, if at least one of said adjustments of said subblocks is above said quantization step.

20. The system of claim 17, wherein said adjustments in said computing means is computed by minimizing a fitting function.

21. A computer-readable medium having stored thereon a plurality of instructions, the plurality of instructions including instructions which, when executed by a processor, cause the processor to perform the steps comprising of:

(a) selecting a block of decoded pixel values;

(b) dividing said block into a plurality of subblocks;

(c) computing an adjustment for each of said subblocks;

(d) applying a forward transform to said plurality of adjustments to produce a plurality of transform coefficients;

(e) determining whether said plurality of adjustments should be applied to said decoded pixel values; and

(f) adjusting said decoded pixel values within each subblock in accordance with said plurality of adjustments, if said determining step (e) determines that said plurality of adjustments should be applied.