Image encoding and decoding method, image processing device and computer storage medium

• US 10,575,022 B2
• Filed: 06/06/2016
• Issued: 02/25/2020
• Est. Priority Date: 06/09/2015
• Status: Active Grant

• Current Assignee: Tongji University, ZTE Corporation
• Original Assignee: ZTE Corporation
• Inventors: Lin, Tao, Li, Ming, Wu, Ping, Shang, Guoqiang, Wu, Zhao
• Primary Examiner(s): Dang, Philip P.
• Application Number:
  US15/580,716
• Publication Number:
  US 20180205971A1
• Time in Patent Office: 1,359 Days
• CPC Class Codes: H03M 13/27 using interleaving techniques
• CPC Class Codes: H03M 7/40 Conversion to or from variable length codes, e.g. Shannon-Fano code, Huffman code, Morse code
• CPC Class Codes: H04N 19/13 Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
• CPC Class Codes: H04N 19/157 Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
• CPC Class Codes: H04N 19/182 the unit being a pixel
• CPC Class Codes: H04N 19/184 the unit being bits, e.g. of the compressed video stream
• CPC Class Codes: H04N 19/593 involving spatial prediction techniques
• CPC Class Codes: H04N 19/91 Entropy coding, e.g. variable length coding [VLC] or arithmetic coding
• CPC Class Codes: H04N 19/96 Tree coding, e.g. quad-tree coding

Abstract

Disclosed are an image encoding and decoding method, image processing device, and computer storage medium, comprising: entropy-encoding a plurality of replication parameters of a current encoding block according to their order after adjustment, to generate a video code stream comprising information of the plurality of replication parameters; said plurality of replication parameters comprising one or more types of replication parameter components. Parsing the video code stream, comprising the information of the plurality of replication parameters, of a decoding block; entropy-decoding said plurality of replication parameters to obtain binary code of said plurality of replication parameters; adjusting said binary code to obtain the values of said plurality of replication parameters; said plurality of replication parameters comprising one or more types of replication parameter components.

Claims

1. An image coding method, comprising:

performing a regulation processing on a plurality of copying parameters of a current coding block; and

performing an entropy coding on the plurality of the copying parameters according to an order of the plurality of the copying parameters that results from the regulation processing, and generating a video bitstream comprising the plurality of the copying parameters,

wherein the plurality of the copying parameters comprise one or more types of copying parameter components,

wherein the one or more types of copying parameter components comprises at least one of a displacement vector horizontal component, a displacement vector vertical component, a one-dimensional displacement vector, a linear address, a relative linear address, an index, a palette linear address, a relative index, a palette relative linear address, a copying length, a copying width, a copying height, a rectangle width, a rectangle length, or an unmatched pixel,

wherein performing the regulation processing on the plurality of the copying parameters of the current coding block comprises;

dividing the plurality of the copying parameters into N subparts according to a predetermined rule, N≥

2; and

classifying, merging and re-sequencing same subparts of the plurality of the copying parameters; and

then performing the entropy coding on the same subparts sequentially,

wherein each of the N subparts of the plurality of the copying parameters is null or non-null.

2. The image coding method according to claim 1, wherein the current coding block is a coding region of an image, and the current coding block comprises at least one of:

a Largest Coding Unit (LCU), a Coding Tree Unit (CTU), a Coding Unit (CU), a sub-region of the CU, a Prediction Unit (PU) or a Transform Unit (TU).

3. The image coding method according to claim 1, wherein performing the regulation processing on the plurality of the copying parameters of the current coding block further comprises:

dividing the plurality of the copying parameters into bin parts with a Context Model (CM) and bin parts without the CM.

4. The image coding method according to claim 3, wherein a copying manner used for the current coding block is a pixel string copying manner combined with a palette, and

one or more types of following copying parameter components comprised in the plurality of the copying parameters are divided into the bin parts with the CM and the bin parts without the CM:

a type of a pixel segment, a copying position component 1 of the pixel segment, a copying position component 2 of the pixel segment, a copying position component 3 of the pixel segment and a copying length of the pixel segment,

wherein the pixel segment is a partition of the current coding block and comprises a copying parameter of the plurality of the copying parameters,

wherein the pixel string copying manner performs an operation to copy a reference string from a reconstructed reference pixel sample set for a coding string in the current block and assign a numerical value of the reference string.

5. The image coding method according to claim 3, wherein a copying manner used for the current coding block is a pixel string copying manner combined with a palette;

and

when the plurality of the copying parameters are divided into the bin parts with the CM and the bin parts without the CM, the bin parts with the CM comprise one or more of following bins;

a type t of a pixel segment, wherein it is indicated that a reference pixel is from the palette when t adopts a first value; and

it is indicated that the reference pixel is not from the palette when t adopts a second value;

an ABOVE copying position a of the pixel segment, wherein a exists when t indicates that the reference pixel is not from the palette, it is indicated that the reference pixel is at the ABOVE position when a adopts a third value, and it is indicated that the reference pixel is not at the ABOVE position when a adopts a fourth value;

a displacement vector horizontal component x0 of the pixel segment, wherein x0 exists when a indicates that the reference pixel is not at the ABOVE position, it is indicated that the displacement vector horizontal component of the pixel is zero when x0 adopts a fifth value, and it is indicated that the displacement vector horizontal component of the pixel segment is not zero when x0 adopts a sixth value;

a displacement vector vertical component y0 of the pixel segment, wherein y0 exists when a indicates that the reference pixel is not at the ABOVE position, it is indicated that the displacement vector vertical component of the pixel segment is zero when y0 adopts a seventh value, and it is indicated that the displacement vector vertical component of the pixel segment is not zero when y0 adopts an eighth value; and

a first bin b₀ in a binary representation of a copying length of the pixel segment,

wherein the pixel segment is a partition of the current coding block and comprises a copying parameter of the plurality of the copying parameters,

wherein the pixel string copying manner performs an operation to copy a reference string from a reconstructed reference pixel sample set for a coding string in the current block and assign a numerical value of the reference string.

6. The image coding method according to claim 3, further comprising:

when the plurality of the copying parameters are divided into the bin parts with the CM and the bin parts without the CM, classifying the bin parts with the CM of the plurality of the copying parameters into a group; and

classifying the bin parts without the CM of the plurality of the copying parameters into another group,

wherein the entropy coding is sequentially performed on two groups.

7. The image coding method according to claim 3, further comprising:

when the plurality of the copying parameters are divided into the bin parts with the CM and the bin parts without the CM, performing the entropy coding on the copying parameter components of a same type in the bin parts with the CM in a centralized manner such that the entropy coding is performed sequentially, and performing the entropy coding on the copying parameter components of a same type in the bin parts without the CM in the centralized manner.

8. The image coding method according to claim 3, further comprising:

when the plurality of the copying parameters are divided into the bin parts with the CM and the bin parts without the CM, performing the entropy coding on all bins of the plurality of the copying parameters with the copying parameter components of a same type in the bin parts with the CM in a centralized manner such that the entropy coding is performed sequentially, and performing the entropy coding on all bins of the plurality of the copying parameters with the copying parameter components of a same type in the bin parts without the CM in the centralized manner.

9. An image decoding method, comprising:

parsing a video bitstream comprising a plurality of copying parameters, of a decoding block;

performing an entropy decoding on the plurality of the copying parameters to obtain bins of the plurality of the copying parameters; and

performing a regulation operation on the bins to obtain values of the plurality of the copying parameters,

wherein the plurality of the copying parameters comprise one or more types of the copying parameter components;

wherein the one or more types of copying parameter components comprises at least one of a displacement vector horizontal component, a displacement vector vertical component, a one-dimensional displacement vector, a linear address, a relative linear address, an index, a palette linear address, a relative index, a palette relative linear address, a copying length, a copying width, a copying height, a rectangle width, a rectangle length, or an unmatched pixel;

wherein performing the regulation operation on the bins to obtain the values of the plurality of the copying parameters comprises;

dividing the bins into N subparts according to a predetermined rule, N≥

2; and

classifying, merging and mapping bins in the N subparts of the bins into reconstructed values of the plurality of the copying parameters,

wherein each of the N subparts of the plurality of the copying parameters is null or non-null.

10. The image decoding method according to claim 9, wherein the decoding block is a decoding region of an image, and the decoding block comprises at least one of:

a Largest Coding Unit (LCU), a Coding Tree Unit (CTU), a Coding Unit (CU), a sub-region of the CU, a Prediction Unit (PU) or a Transform Unit (TU).

11. The image decoding method according to claim 9, further comprising:

dividing the bins into bin parts with the entropy decoding based on a Context Model (CM) and bin parts without the entropy decoding based on the CM;

extracting corresponding bins from at least one of the bin parts with the entropy decoding based on the CM or the bin parts without the entropy decoding with the CM according to a corresponding table about the CM for the bins of the plurality of the copying parameters in an entropy decoding process, to form the bins of the plurality of the copying parameters; and

converting the bins of the plurality of the copying parameters into the values of the plurality of the copying parameters.

12. The image decoding method according to claim 11, wherein a copying manner used for the decoding block is a pixel string copying manner combined with a palette;

and

the bins corresponding to following one or more types of the copying parameters are divided into the bin parts with the CM and the bin parts without the CM, comprising;

a type of a pixel segment, a copying position component 1 of the pixel segment, a copying position component 2 of the pixel segment, a copying position component 3 of the pixel segment and a copying length of the pixel segment,

wherein the pixel segment is a partition of the current coding block and comprises a copying parameter of the plurality of the copying parameters,

wherein the pixel string copying manner performs an operation to copy a reference string from a reconstructed reference pixel sample set for a coding string in the current block and assign a numerical value of the reference string.

13. The image decoding method according to claim 11, wherein a copying manner used for the decoding block is a pixel string copying manner combined with a palette, and

when the bins of the plurality of the copying parameters are divided into the bin parts with the CM and the bin parts without the CM, the bin parts with the CM comprise the following one or more bins:

a type t of a pixel segment, wherein it is indicated that a reference pixel is from the palette when t adopts a first value, and it is indicated that the reference pixel is not from the palette when t adopts a second value;

an ABOVE copying position a of the pixel segment, wherein a exists when t indicates that the reference pixel is not from the palette, it is indicated that the reference pixel is at the ABOVE position when a adopts a third value, and it is indicated that the reference pixel is not at the ABOVE position when a adopts a fourth value;

a displacement vector horizontal component x0 of the pixel segment, wherein x0 exists when a indicates that the reference pixel is not at the ABOVE position, it is indicated that the displacement vector horizontal component of the pixel segment is zero when x0 adopts a fifth value, and it is indicated that the displacement vector horizontal component of the pixel segment is not zero when x0 adopts a sixth value;

a displacement vector vertical component y0 of the pixel segment, wherein y0 exists when a indicates that the reference pixel is not at the ABOVE position, it is indicated that the displacement vector vertical component of the pixel segment is zero when y0 adopts a seventh value, and it is indicated that the displacement vector vertical component of the pixel segment is not zero when y0 adopts an eighth value; and

a first bin b₀ in a binary representation of a copying length of the pixel segment,

wherein the pixel segment is a partition of the current coding block and comprises a copying parameter of the plurality of the copying parameters,

wherein the pixel string copying manner performs an operation to copy a reference string from a reconstructed reference pixel sample set for a coding string in the current block and assign a numerical value of the reference string.

14. The image decoding method according to claim 11, further comprising:

when the bins of the plurality of the copying parameters are divided into the bin parts with the CM and the bin parts without the CM, sequentially performing the entropy decoding on the video bitstream corresponding to the bins with the CM of the plurality of the copying parameters, and performing the entropy decoding on the video bitstream corresponding to the bin parts without the CM of the plurality of the copying parameters;

extracting the corresponding bins from at least one of the bin parts with the entropy decoding based on the CM or the bin parts without the entropy decoding based on the CM according to the corresponding table about the CM for the bins of the plurality of the copying parameters in the entropy decoding process to form the bins of the plurality of the copying parameters; and

converting the bins of the plurality of the copying parameters into the values of the plurality of the copying parameters.

15. The image decoding method according to claim 11, further comprising:

during the entropy decoding, sequentially performing the entropy decoding on a group formed by the bin parts with the CM in the bins of the plurality of the copying parameters and another group formed by the bin parts without the CM in the bins of the plurality of the copying parameters;

when the bins of the plurality of the copying parameters are divided into the bin parts with the CM and the bin parts without the CM, performing the entropy decoding on a bitstream corresponding to the bins of the copying parameter components of a same type in the bin parts with the CM, and correspondingly performing the entropy decoding on the bins of the copying parameter components of a same type in the bin parts without the CM;

extracting the corresponding bins from at least one of the bin parts with the entropy decoding based on the CM or the bin parts without the entropy decoding based on the CM according to the corresponding table about the CM for the bins of the plurality of the copying parameters in the entropy decoding process to form the bins of the plurality of the copying parameters; and

converting the bins of the plurality of the copying parameters into the values of the plurality of the copying parameters.

16. The image decoding method according to claim 11, further comprising:

when the bins of the plurality of the copying parameters are divided into the bin parts with the CM and the bin parts without the CM, performing the entropy decoding on the video bitstream corresponding to the bins of the plurality of the copying parameters with the copying parameter components of a same type in the bin parts with the CM, and performing the entropy decoding on the video bitstream corresponding to the bins of the plurality of the copying parameters with the copying parameter components of a same type in the bin parts without the CM;

extracting the corresponding bins from at least one of the bin parts with the entropy decoding based on the CM or the bin parts without the entropy decoding based on the CM according to the corresponding table about the CM for the bins of the plurality of the copying parameters in the entropy decoding process to form the bins of the plurality of the copying parameters; and

converting the bins of the plurality of the copying parameters into the values of the plurality of the copying parameters.

17. An image processing equipment, comprising a processor and a memory for storing a set of instructions executable for the processor, wherein when the instructions are executed by the processor, the processor is arranged to:

perform a regulation processing on a plurality of copying parameters of a current coding block; and

perform an entropy coding on the plurality of the copying parameters according to an order of the plurality of the copying parameters that results from the regulation processing, and generate a video bitstream comprising the plurality of the copying parameters,

wherein the plurality of the copying parameters comprise one or more types of copying parameter components,

wherein the one or more types of copying parameter components comprises at least one of a displacement vector horizontal component, a displacement vector vertical component, a one-dimensional displacement vector, a linear address, a relative linear address, an index, a palette linear address, a relative index, a palette relative linear address, a copying length, a copying width, a copying height, a rectangle width, a rectangle length, or an unmatched pixel, and

wherein the processor is further arranged to;

divide the plurality of the copying parameters into N subparts according to a predetermined rule, N≥

2; and

classify, merge and re-sequence same subparts of the plurality of the copying parameters; and

then perform the entropy coding on the same subparts sequentially,

wherein each of the N subparts of the plurality of the copying parameters is null or non-null.

18. An image processing equipment, comprising a processor and a memory for storing a set of instructions executable for the processor, wherein when the instructions are executed by the processor, the processor is arranged to:

parse a video bitstream comprising a plurality of copying parameters, of a decoding block; and

perform an entropy decoding on the plurality of the copying parameters to obtain bins of the plurality of the copying parameters, and perform a regulation operation on the bins to obtain values of the plurality of the copying parameters,

wherein the plurality of the copying parameters comprise one or more types of copying parameter components,

wherein the one or more types of copying parameter components comprises at least one of a displacement vector horizontal component, a displacement vector vertical component, a one-dimensional displacement vector, a linear address, a relative linear address, an index, a palette linear address, a relative index, a palette relative linear address, a copying length, a copying width, a copying height, a rectangle width, a rectangle length, or an unmatched pixel, and

wherein the processor is further arranged to;

divide the bins into N subparts according to a predetermined rule, N≥

2; and

classify, merge and map bins in the N subparts of the bins into reconstructed values of the plurality of the copying parameters,

wherein each of the N subparts of the plurality of the copying parameters is null or non-null.