Big data node collaborative processing system

By using multiple data nodes to cooperate in big data processing, reference coding of color components is performed based on the difference value of cyan components, the problem of excessive intra-coded data and calculation amount in big data processing is solved, and the encoding is improved efficiency.

CN118524218BActive Publication Date: 2024-11-29SHANDONG FEIXIANG INFORMATION TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202410822234.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-25
Publication Date
2024-11-29
Estimated Expiration
2044-06-25

AI Technical Summary

Technical Problem

In the current big data processing, the intra-coding of each color component is carried out independently, and the reference and utilization of mutual data cannot be achieved, resulting in the amount of data and calculations of intra-coding remain high.

Method used

By cooperating with the first big data node, the second big data node and the third big data node, the encoded cyan component is different from the target image block by pixel on the left and top of the target image block, respectively. The color difference value chunking is obtained, and the chunk with the least non-zero data is used as the reference code chunking of other color components, completing the intra-coding of magenta, yellow and black components.

Benefits of technology

The encoding speed and efficiency are improved, and the processing volume of uncoded color components is reduced through the filtering of reference encoding blocks, and the efficiency of big data processing is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118524218B_ABST
    Figure CN118524218B_ABST
Patent Text Reader

Abstract

The present invention relates to a large data node collaborative processing system, comprising: a first large data node, used to determine a first candidate block and a second candidate block of a target image block to be encoded according to the position of the target image block to be encoded in the current image frame, and the cyan component of the target image block has completed intra-frame encoding; a second large data node and a third large data node, used to determine a reference coding block for performing intra-frame encoding of other color components based on the intra-frame encoding result of the cyan component. Through the present invention, the reference coding block of other color components can be determined based on the intra-frame encoding result of the encoded cyan component, and the first large data node, the second large data node and the third large data node are used to collaboratively complete the numerical analysis of the reference coding block for performing intra-frame encoding of the magenta component, the yellow component and the black component of the target image block to be encoded, thereby improving the speed and efficiency of intra-frame encoding.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of big data processing, and in particular to a big data node collaborative processing system. Background Art

[0002] Big data refers to data that is large and complex. Big data processing is usually the collection and manipulation of data items to produce meaningful information. In this sense, it can be considered a subset of information processing, which detects changes in information in any way.

[0003] Information is growing at an ever-increasing rate and scale. The world's effective capacity to exchange information via telecommunications networks was 281PB in 1986, 471PB in 1993, 2.2 exabytes in 2000, 65 exabytes in 2007, and Internet traffic reached 2,667 exabytes per year in 2023. According to one estimate, one-third of the world's stored information is in the form of alphanumeric text and still image data, the most useful format for most big data applications. This also shows the potential for data that has not yet been used (i.e., in the form of video and audio content).

[0004] The application field of big data processing is very wide. For example, the characteristics of big data processing can be used to perform intra-frame coding processing on each color component of the target image block to be intra-frame coded. However, the intra-frame coding processing of each color component is currently performed independently, and the mutual data reference and utilization cannot be achieved, resulting in the current intra-frame coding data volume and computational complexity remaining high. Summary of the invention

[0005] In order to solve the technical problems in the related fields, the present invention provides a big data node collaborative processing system, which can perform pixel-by-pixel difference of the cyan component between the left image block and the top image block of the target image block based on the encoded cyan component to obtain two color difference blocks, and use the image block corresponding to the color difference block with the least number of non-zero data as the reference coding block for intra-frame encoding of other color components of the target image block, such as magenta component, yellow component and black component, and use the first big data node, the second big data node and the third big data node to collaborate to complete the numerical analysis of the reference coding block for intra-frame encoding of the magenta component, yellow component and black component of the target image block to be encoded, thereby completing the targeted screening of the reference coding blocks of the unencoded remaining color components based on big data processing.

[0006] According to the present invention, a big data node collaborative processing system is provided, the system comprising:

[0007] A first large data node is used to obtain a target image block to be encoded, the target image block is located in a current image frame, and a first candidate block and a second candidate block of the target image block are determined according to the position of the target image block in the current image frame, and a cyan component of the target image block has completed intra-frame encoding;

[0008] A second large data node, connected to the first large data node, for parsing a first color difference block obtained by performing a pixel-by-pixel difference between a color block corresponding to the cyan component of the first candidate block and a color block corresponding to the cyan component of the target image block, and a second color difference block obtained by performing a pixel-by-pixel difference between a color block corresponding to the cyan component of the second candidate block and a color block corresponding to the cyan component of the target image block;

[0009] a third large data node, connected to the second large data node, for obtaining the number of non-zero cyan component difference values ​​in the first color difference value block as a first reference number, and obtaining the number of non-zero cyan component difference values ​​in the second color difference value block as a second reference number, and using the reference block corresponding to the smallest number of the first reference number and the second reference number as a reference coding block for performing intra-frame coding on the magenta component, the yellow component and the black component of the target image block;

[0010] Wherein, obtaining a target image block to be encoded, the target image block is located in a current image frame, and determining a first candidate block and a second candidate block of the target image block according to a position of the target image block in the current image frame comprises: the first candidate block of the target image block is an image block located to the left of the target image block in the current image frame, and the second candidate block of the target image block is an image block located directly above the target image block in the current image frame;

[0011] Wherein, the first color difference block obtained by analyzing the pixel-by-pixel difference between the color block corresponding to the cyan component of the first candidate block and the color block corresponding to the cyan component of the target image block, and the second color difference block obtained by analyzing the pixel-by-pixel difference between the color block corresponding to the cyan component of the second candidate block and the color block corresponding to the cyan component of the target image block include: the color block corresponding to the cyan component of the first candidate block is a color block composed of the cyan components corresponding to the constituent pixels of the first candidate block, and the color block corresponding to the cyan component of the second candidate block is a color block composed of the cyan components corresponding to the constituent pixels of the second candidate block;

[0012] Among them, the first color difference block obtained by analyzing the pixel-by-pixel difference between the color block corresponding to the cyan component of the first candidate block and the color block corresponding to the cyan component of the target image block, and the second color difference block obtained by analyzing the pixel-by-pixel difference between the color block corresponding to the cyan component of the second candidate block and the color block corresponding to the cyan component of the target image block also include: the cyan component is in the CMYK color space and the CMYK color space includes a cyan component, a magenta component, a yellow component and a black component.

[0013] The present invention has at least the following three important inventive features:

[0014] Invention point A: The first large data node, the second large data node and the third large data node cooperate to complete the numerical analysis of the reference coding block for performing intra-frame coding on the magenta component, the yellow component and the black component of the target image block to be coded, thereby improving the coding speed and efficiency;

[0015] Invention point B: Based on the coded cyan component, the left image block and the top image block of the target image block are respectively subjected to pixel-by-pixel difference of the cyan component with the target image block to obtain two color difference blocks, and the image block corresponding to the color difference block with the least amount of non-zero data is used as the reference coding block for performing intra-frame coding on the magenta component, yellow component and black component of the target image block, thereby completing the targeted screening of the reference coding blocks of the remaining uncoded color components;

[0016] Invention point C: Specifically, when parsing the first color difference block obtained by making a pixel-by-pixel difference between the color block corresponding to the cyan component of the left image block and the color block corresponding to the cyan component of the target image block, and parsing the second color difference block obtained by making a pixel-by-pixel difference between the color block corresponding to the cyan component of the top image block and the color block corresponding to the cyan component of the target image block, the color block corresponding to the cyan component of the left candidate block is a color block composed of the cyan components corresponding to the respective pixels of the left candidate block, and the color block corresponding to the cyan component of the top candidate block is a color block composed of the cyan components corresponding to the respective pixels of the top candidate block. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein:

[0018] Figure 1 Schematic diagram of the internal structure of a large data node collaborative processing system according to implementation A of the present invention.

[0019] Figure 2 Schematic diagram of the internal structure of a large data node collaborative processing system according to implementation B of the present invention.

[0020] Figure 3 Schematic diagram of the internal structure of a large data node collaborative processing system according to implementation scheme C of the present invention. DETAILED DESCRIPTION

[0021] The implementation scheme of the big data node collaborative processing system of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Figure 1 This is a schematic diagram of the internal structure of a big data node collaborative processing system according to Embodiment A of the present invention, wherein the system comprises:

[0023] A first large data node is used to obtain a target image block to be encoded, the target image block is located in a current image frame, and a first candidate block and a second candidate block of the target image block are determined according to the position of the target image block in the current image frame, and a cyan component of the target image block has completed intra-frame encoding;

[0024] Alternatively, a first cloud computing node may be selected to replace the first data node, and used to obtain a target image block to be encoded, the target image block being located in a current image frame, and a first candidate block and a second candidate block of the target image block being determined according to a position of the target image block in the current image frame, wherein a cyan component of the target image block has completed intra-frame encoding;

[0025] A second large data node, connected to the first large data node, for parsing a first color difference block obtained by performing a pixel-by-pixel difference between a color block corresponding to the cyan component of the first candidate block and a color block corresponding to the cyan component of the target image block, and a second color difference block obtained by performing a pixel-by-pixel difference between a color block corresponding to the cyan component of the second candidate block and a color block corresponding to the cyan component of the target image block;

[0026] a third large data node, connected to the second large data node, for obtaining the number of non-zero cyan component difference values ​​in the first color difference value block as a first reference number, and obtaining the number of non-zero cyan component difference values ​​in the second color difference value block as a second reference number, and using the reference block corresponding to the smallest number of the first reference number and the second reference number as a reference coding block for performing intra-frame coding on the magenta component, the yellow component and the black component of the target image block;

[0027] Likewise, the second cloud computing node and the third cloud computing node may be selected to replace the second data node and the third data node respectively;

[0028] Wherein, obtaining a target image block to be encoded, the target image block is located in a current image frame, and determining a first candidate block and a second candidate block of the target image block according to a position of the target image block in the current image frame comprises: the first candidate block of the target image block is an image block located to the left of the target image block in the current image frame, and the second candidate block of the target image block is an image block located directly above the target image block in the current image frame;

[0029] Wherein, the first color difference block obtained by analyzing the pixel-by-pixel difference between the color block corresponding to the cyan component of the first candidate block and the color block corresponding to the cyan component of the target image block, and the second color difference block obtained by analyzing the pixel-by-pixel difference between the color block corresponding to the cyan component of the second candidate block and the color block corresponding to the cyan component of the target image block include: the color block corresponding to the cyan component of the first candidate block is a color block composed of the cyan components corresponding to the constituent pixels of the first candidate block, and the color block corresponding to the cyan component of the second candidate block is a color block composed of the cyan components corresponding to the constituent pixels of the second candidate block;

[0030] Wherein, the first color difference block obtained by performing pixel-by-pixel difference analysis on the color block corresponding to the cyan component of the first candidate block and the color block corresponding to the cyan component of the target image block, and the second color difference block obtained by performing pixel-by-pixel difference analysis on the color block corresponding to the cyan component of the second candidate block and the color block corresponding to the cyan component of the target image block further include: the cyan component is in the CMYK color space and the CMYK color space includes a cyan component, a magenta component, a yellow component and a black component;

[0031] Wherein, obtaining a target image block to be encoded, the target image block being located in a current image frame, and determining a first candidate block and a second candidate block of the target image block according to a position of the target image block in the current image frame further comprises: the target image block, the first candidate block and the second candidate block have the same size;

[0032] And wherein, the target image block, the first candidate block and the second candidate block have the same size includes: the target image block, the first candidate block and the second candidate block are all square image blocks.

[0033] Figure 2 Schematic diagram of the internal structure of a large data node collaborative processing system according to implementation B of the present invention.

[0034] Compared with implementation scheme A, the big data node collaborative processing system shown in implementation scheme B may also include the following components:

[0035] An ASIC control device, connected to the first large data node, the second large data node, and the third large data node, respectively, and used to provide configuration operations of working parameters for the first large data node, the second large data node, and the third large data node in a time-sharing manner;

[0036] The ASIC control device is connected to the first large data node, the second large data node, and the third large data node respectively, and is used to provide the first large data node, the second large data node, and the third large data node with configuration operations of working parameters in a time-sharing manner, including: the first large data node, the second large data node, and the third large data node share the same working parameter configuration interface;

[0037] Wherein, the first large data node, the second large data node, and the third large data node use different configuration address data;

[0038] Wherein, a parallel connection of a communication data link is established between each of the first large data node, the second large data node, and the third large data node through a parallel data bus;

[0039] Among them, a parallel connection of a communication data link established between each of the first large data node, the second large data node, and the third large data node through a parallel data bus includes: the parallel data bus is one of an 8-bit parallel data bus, a 16-bit parallel data bus, and a 32-bit parallel data bus.

[0040] Figure 3 Schematic diagram of the internal structure of a large data node collaborative processing system according to implementation scheme C of the present invention.

[0041] Compared with implementation scheme A, the big data node collaborative processing system shown in implementation scheme C may also include the following components:

[0042] a power supply processing device, connected to the first large data node, the second large data node, and the third large data node, respectively, and used to provide power distribution support of different working voltages for the first large data node, the second large data node, and the third large data node in a time-sharing manner;

[0043] The power supply processing device is respectively connected to the first large data node, the second large data node, and the third large data node, and is used to provide power distribution support with different working voltages for the first large data node, the second large data node, and the third large data node in a time-sharing manner, including: two or more devices with the same working voltage requirements in the first large data node, the second large data node, and the third large data node use the same power supply line;

[0044] The power supply processing device is connected to the first large data node, the second large data node, and the third large data node respectively, and is used to provide power distribution support with different working voltages for the first large data node, the second large data node, and the third large data node in a time-sharing manner, and further includes: the power supply processing device is an uninterruptible power supply device;

[0045] Among them, the power supply processing device is connected to the first large data node, the second large data node, and the third large data node respectively, and is used to provide power distribution support with different working voltages for the first large data node, the second large data node, and the third large data node in time-sharing, and further includes: providing different working voltages including 3.3V working voltage for the first large data node, the second large data node, and the third large data node in time-sharing;

[0046] And wherein, the power supply processing device is connected to the first large data node, the second large data node, and the third large data node, respectively, and is used to provide power distribution support with different working voltages for the first large data node, the second large data node, and the third large data node in a time-sharing manner, and also includes: providing different working voltages including a 5V working voltage for the first large data node, the second large data node, and the third large data node in a time-sharing manner.

[0047] In addition, in the big data node collaborative processing system, a target image block to be encoded is obtained, the target image block is located in a current image frame, and a first candidate block and a second candidate block of the target image block are determined according to the position of the target image block in the current image frame, further comprising: the target image block, the first candidate block and the second candidate block have the same size and the size is monotonically inversely correlated with the signal-to-noise ratio of the current image frame.

[0048] Among them, the target image block, the first candidate block and the second candidate block have the same size and the size is monotonically inversely correlated with the signal-to-noise ratio of the current image frame, including: the higher the signal-to-noise ratio of the current image frame, the smaller the size.

[0049] The big data node collaborative processing system of the present invention is used to address the technical problems in the prior art of intra-frame coding based on big data processing that the amount of data and the amount of calculation are high. The reference coding blocks of other color components are determined based on the intra-frame coding results of the encoded cyan component, and the first big data node, the second big data node, and the third big data node are used in a collaborative manner to complete the numerical analysis of the reference coding blocks for intra-frame coding of the magenta component, the yellow component, and the black component of the target image block to be encoded, thereby solving the above-mentioned technical problems.

[0050] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A big data node collaborative processing system, characterized in that: The system comprises: A first large data node is used to obtain a target image block to be encoded, the target image block is located in a current image frame, and a first candidate block and a second candidate block of the target image block are determined according to the position of the target image block in the current image frame, and a cyan component of the target image block has completed intra-frame encoding; A second large data node, connected to the first large data node, for parsing a first color difference block obtained by performing a pixel-by-pixel difference between a color block corresponding to the cyan component of the first candidate block and a color block corresponding to the cyan component of the target image block, and a second color difference block obtained by performing a pixel-by-pixel difference between a color block corresponding to the cyan component of the second candidate block and a color block corresponding to the cyan component of the target image block; a third large data node, connected to the second large data node, for obtaining the number of non-zero cyan component difference values ​​in the first color difference value block as a first reference number, and obtaining the number of non-zero cyan component difference values ​​in the second color difference value block as a second reference number, and using the color difference value block corresponding to the smallest number of the first reference number and the second reference number as a reference coding block for performing intra-frame coding on the magenta component, the yellow component and the black component of the target image block; Wherein, obtaining a target image block to be encoded, the target image block is located in a current image frame, and determining a first candidate block and a second candidate block of the target image block according to a position of the target image block in the current image frame comprises: the first candidate block of the target image block is an image block located to the left of the target image block in the current image frame, and the second candidate block of the target image block is an image block located directly above the target image block in the current image frame; Wherein, the first color difference block obtained by analyzing the pixel-by-pixel difference between the color block corresponding to the cyan component of the first candidate block and the color block corresponding to the cyan component of the target image block, and the second color difference block obtained by analyzing the pixel-by-pixel difference between the color block corresponding to the cyan component of the second candidate block and the color block corresponding to the cyan component of the target image block include: the color block corresponding to the cyan component of the first candidate block is a color block composed of the cyan components corresponding to the constituent pixels of the first candidate block, and the color block corresponding to the cyan component of the second candidate block is a color block composed of the cyan components corresponding to the constituent pixels of the second candidate block; Among them, the first color difference block obtained by analyzing the pixel-by-pixel difference between the color block corresponding to the cyan component of the first candidate block and the color block corresponding to the cyan component of the target image block, and the second color difference block obtained by analyzing the pixel-by-pixel difference between the color block corresponding to the cyan component of the second candidate block and the color block corresponding to the cyan component of the target image block also include: the cyan component is in the CMYK color space and the CMYK color space includes a cyan component, a magenta component, a yellow component and a black component.

2. The big data node collaborative processing system according to claim 1, characterized in that: Acquiring a target image block to be encoded, the target image block being located in a current image frame, and determining a first candidate block and a second candidate block of the target image block according to a position of the target image block in the current image frame further comprises: the target image block, the first candidate block and the second candidate block have the same size; The target image block, the first candidate block and the second candidate block having the same size include: the target image block, the first candidate block and the second candidate block are all square image blocks.

3. The big data node collaborative processing system according to claim 2, characterized in that: The system further comprises: An ASIC control device, connected to the first large data node, the second large data node, and the third large data node, respectively, for providing configuration operations of working parameters for the first large data node, the second large data node, and the third large data node in a time-sharing manner; Among them, the ASIC control device is connected to the first large data node, the second large data node, and the third large data node respectively, and is used to provide the configuration operation of working parameters for the first large data node, the second large data node, and the third large data node in a time-sharing manner, including: the first large data node, the second large data node, and the third large data node share the same working parameter configuration interface.

4. The big data node collaborative processing system according to claim 3, characterized in that: The first large data node, the second large data node, and the third large data node use different configuration address data.

5. The big data node collaborative processing system according to claim 3, characterized in that: Establishing parallel connections of communication data links between the first large data node, the second large data node, and the third large data node through parallel data buses; Among them, the parallel connection of the communication data link established between each of the first large data node, the second large data node, and the third large data node through a parallel data bus includes: the parallel data bus is one of an 8-bit parallel data bus, a 16-bit parallel data bus, and a 32-bit parallel data bus.

6. The big data node collaborative processing system according to claim 2, characterized in that: The system further comprises: a power supply processing device, connected to the first large data node, the second large data node, and the third large data node, respectively, and used to provide power distribution support of different working voltages for the first large data node, the second large data node, and the third large data node in a time-sharing manner; Among them, the power supply processing device is connected to the first large data node, the second large data node, and the third large data node respectively, and is used to provide power distribution support with different working voltages for the first large data node, the second large data node, and the third large data node in a time-sharing manner, including: among the first large data node, the second large data node, and the third large data node, two or more devices with the same working voltage requirements use the same power supply line.

7. The big data node collaborative processing system according to claim 6, characterized in that: A power supply processing device is connected to the first large data node, the second large data node, and the third large data node, respectively, and is used to provide power distribution support with different working voltages for the first large data node, the second large data node, and the third large data node in a time-sharing manner, and also includes: the power supply processing device is an uninterruptible power supply device.

8. The big data node collaborative processing system according to claim 7, characterized in that: The power supply processing device is connected to the first large data node, the second large data node, and the third large data node respectively, and is used to provide power distribution support with different working voltages for the first large data node, the second large data node, and the third large data node in a time-sharing manner, and also includes: providing different working voltages including a 3.3V working voltage for the first large data node, the second large data node, and the third large data node in a time-sharing manner.

9. The big data node collaborative processing system according to claim 8, characterized in that: The power supply processing device is connected to the first large data node, the second large data node, and the third large data node respectively, and is used to provide power distribution support with different working voltages for the first large data node, the second large data node, and the third large data node in a time-sharing manner, and also includes: providing different working voltages including a 5V working voltage for the first large data node, the second large data node, and the third large data node in a time-sharing manner.

Citation Information

Patent Citations

  • Intra-frame prediction method and device and coding and decoding method and device based on intra-frame prediction

    CN112422968A

  • Intra-frame prediction method, video / image coding and decoding method and related device

    CN113489974A