Data block encoding method and apparatus

Patent metadata

Field	Value
Publication number	US-11019343-B2
Application number	US-201715660737-A
Country	US
Kind code	B2
Filing date	Jul 26, 2017
Priority date	Jul 28, 2016
Publication date	May 25, 2021
Grant date	May 25, 2021

Abstract

A data block encoding method and apparatus are provided. The data block encoding method includes: determining whether a data block includes an intensive edge region, where the intensive edge region is a region including an image distortion generated by ring noise, and the data block is a data block in a to-be-encoded image; and when the data block includes an intensive edge region, reducing a value of a quantization parameter used for encoding the data block, and encoding the data block by using a reduced value of the quantization parameter, or dividing the data block into multiple coding units of different sizes according to different coding layers, adjusting rate-distortion costs of the multiple coding units of different sizes, and encoding the data block by using a coding unit with a minimum rate-distortion cost obtained after the adjustment.

First claim

What is claimed is:

1. A data block encoding method, comprising:
determining whether a data block comprises an intensive edge region comprising an image distortion generated by ring noise, wherein the data block is a data block in a to-be-encoded image; and
when the data block comprises an intensive edge region,
dividing the data block into multiple coding units of different sizes according to different coding layers,
adjusting rate-distortion costs of the multiple coding units of different sizes, wherein the adjustment of the rate-distortion cost of a coding unit is proportional to the size of the coding unit and wherein the adjusted rate-distortion cost of a coding unit is based upon a formula that indicates a rate-distortion cost of a coding unit of size N after an adjustment is equal to an adjustment coefficient corresponding to the coding unit of size N multiplied by the rate-distortion cost of the coding unit of size N before the adjustment and wherein the value of the adjustment coefficient is proportional to the size of the coding unit,
encoding the data block by using a coding unit with a minimum rate-distortion cost obtained after the adjustment.

2. The method according to claim 1 , wherein determining whether a data block comprises an intensive edge region comprises:
dividing the data block into A sub data blocks;
obtaining a variance of the data block and a variance of each sub data block in the A sub data blocks; and
when the variance of the data block is greater than a first threshold, and in the variances of the A sub data blocks a maximum variance is greater than k times a minimum variance, determining that the data block comprises an intensive edge region, wherein A is a positive integer greater than or equal to 2, k is an integer, and the first threshold is determined by a bit width used by a pixel value in the to-be-encoded image.

3. The method according to claim 2 , wherein when the variance of the data block is less than or equal to the first threshold, or in the variances of the A sub data blocks the maximum variance is less than or equal to k times the minimum variance, the method further comprises:
dividing each sub data block in the A sub data blocks into B sub-sub data blocks, wherein B is an integer greater than 0;
obtaining the variance of each of the A sub data blocks and a variance of each of the B sub-sub data blocks comprised in each sub data block; and
when the variance of any sub data block is greater than the first threshold, and in the variances of the B sub-sub data blocks comprised in the any sub data block a maximum variance is greater than k times a minimum variance, determining that the data block comprises an intensive edge region.

4. The method according to claim 1 , further comprising reducing a value of a quantization parameter used for encoding the data block comprising:
reducing, according to a first formula, the value of the quantization parameter used for encoding the data block, wherein the first formula is:
QP′=QP −delta,

wherein QP′ indicates a value of the quantization parameter after the adjustment, QP indicates a value of the quantization parameter before the adjustment, delta indicates an adjustment magnitude, and the adjustment magnitude is any integer greater than 0 but less than 15.

5. The method according to claim 1 , wherein adjusting rate-distortion costs of the multiple coding units of different sizes comprises:
adjusting the rate-distortion costs of the multiple coding units of different sizes according to the formula, wherein the formula is:
Rdo Cost N ′=C N ×Rdo Cost N ,

wherein RdoCost N ′ indicates the rate-distortion cost of the coding unit of the size N after the adjustment, C N indicates the adjustment coefficient corresponding to the coding unit of the size N, a value of the adjustment coefficient is proportional to the size of the coding unit, RdoCost N indicates the rate-distortion cost of the coding unit of the size N before the adjustment, and N indicates the size of the coding unit.

6. A data block encoding apparatus, comprising:
a non-transitory computer readable medium configured to store instructions; and
a processor configured to execute the instructions stored in the non-transitory computer readable medium to:
determine whether a data block comprises an intensive edge region comprising an image distortion generated by ring noise, wherein the data block is a data block in a to-be-encoded image; and
when the data block comprises an intensive edge region, divide the data block into multiple coding units of different sizes according to different coding layers, adjust rate-distortion costs of the multiple coding units of different sizes, wherein the adjustment of the rate-distortion cost of a coding unit is proportional to the size of the coding unit, and wherein the adjusted rate-distortion cost of a coding unit is based upon a formula that indicates a rate-distortion cost of a coding unit of size N after an adjustment is equal to an adjustment coefficient corresponding to the coding unit of size N multiplied by the rate-distortion cost of the coding unit of size N before the adjustment and wherein the value of the adjustment coefficient is proportional to the size of the coding unit, and encode the data block by using a coding unit with a minimum rate-distortion cost obtained after the adjustment.

7. The apparatus according to claim 6 , wherein the processor is configured to:
divide the data block into A sub data blocks;
obtain a variance of the data block and a variance of each sub data block in the A sub data blocks; and
when the variance of the data block is greater than a first threshold, and in the variances of the A sub data blocks a maximum variance is greater than k times a minimum variance, determine that the data block comprises an intensive edge region, wherein A is a positive integer greater than or equal to 2, k is an integer, and the first threshold is determined by a bit width used by a pixel value in the to-be-encoded image.

8. The apparatus according to claim 7 , wherein when the variance of the data block is less than or equal to the first threshold, or in the variances of the A sub data blocks, the maximum variance is less than or equal to k times the minimum variance, the processor is further configured to:
divide each sub data block in the A sub data blocks into B sub-sub data blocks, wherein B is an integer greater than 0;
obtain the variance of each of the A sub data blocks and a variance of each sub-sub data block in the B sub-sub data blocks comprised in each sub data block; and
when the variance of any sub data block is greater than the first threshold, and in the variances of the B sub-sub data blocks comprised in the any sub data block a maximum variance is greater than k times a minimum variance, determine that the data block comprises an intensive edge region.

9. The apparatus according to claim 6 , wherein the processor is configured to:
reduce, according to a first formula, a value of a quantization parameter used for encoding the data block, wherein the first formula is:
QP′=QP −delta,

wherein QP′ indicates a value of the quantization parameter after the adjustment, QP indicates a value of the quantization parameter before the adjustment, delta indicates an adjustment magnitude, and the adjustment magnitude is any integer greater than 0 but less than 15.

10. The apparatus according to claim 6 , wherein the processor is configured to:
adjust the rate-distortion costs of the multiple coding units of different sizes according to the formula, wherein the formula is:
Rdo Cost N ′=C N ×Rdo Cost N ,

wherein RdoCost N ′ indicates the

Assignees

• Huawei Tech Co Ltd

Inventors

• Chen Shaolin
• Xiang Xiaohong
• Wu Dongsheng

Classifications

• H04N19/122

  Selection of transform size, e.g. 8x8 or 2x4x8 DCT; Selection of sub-band transforms of varying structure or type · CPC title

• H04N19/18

  the unit being a set of transform coefficients · CPC title

• H04N19/625

  using discrete cosine transform [DCT] · CPC title

• H04N19/117

  Filters, e.g. for pre-processing or post-processing (sub-band filter banks H04N19/635) · CPC title

• H04N19/196

  being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters (processing of motion vectors H04N19/513) · CPC title