Microcomputer algorithm operating system using content parsing

Abstract

The invention relates to a microcomputer algorithm operating system utilizing content analysis, which comprises a data extraction device arranged in the microcomputer algorithm operating system and used for acquiring the signal-to-noise ratio, the pixel value mean square error, the number of pixel rows and the number of pixel columns of an image where a current coding block is located; and the parameter analysis mechanism is used for intelligently analyzing the coding operand consumed by using the reference coding block to execute intra-frame coding on the current coding block according to the image information of the current coding block by adopting a deep convolutional network model. The microcomputer algorithm operating system using content analysis is stable in operation and intelligent in design. Due to the fact that the object remodeling device arranged in the microcomputer algorithm operating system is introduced and used for analyzing the deep convolutional network model which is designed in a targeted mode, the encoding operation amount which needs to be consumed when any reference encoding block is used for executing intra-frame encoding on the current encoding block is predicted, and therefore a reliable basis is provided for selection of the reference encoding block.

Description

Technical Field

[0001] The present invention relates to the field of computer systems, and in particular to a microcomputer algorithm operating system utilizing content parsing. Background Art

[0002] The software of a computer system is divided into system software, support software and application software. System software consists of operating systems, utility programs, compilers, etc. The operating system implements management and control of various software and hardware resources. Utilities are set up for the convenience of users, such as text editing. The function of the compiler is to translate the program written by the user in assembly language or some high-level language into a machine language program that can be executed by the machine. Support software includes interface software, tool software, environment database, etc. It can support the environment of the machine and provide software development tools. Support software can also be considered as part of the system software. Application software is a special program written by the user according to his needs. It runs with the help of system software and support software and is the outermost layer of the software system.

[0003] In the prior art, a computer system needs to adopt a specially designed algorithm to intelligently analyze the different amounts of coding operations required for intra-frame coding of the current coding block using different reference coding blocks, so as to provide a reliable method for selecting the reference coding block with the lowest amount of coding operations. However, the prior art lacks a corresponding solution. Summary of the invention

[0004] In order to solve the technical problems in the prior art, the present invention provides a microcomputer algorithm operating system using content analysis, which can introduce an object reshaping device set in the microcomputer algorithm operating system to analyze a targeted designed deep convolutional network model. In the deep convolutional network model, the deep convolutional network is trained for each time to obtain a deep convolutional network after each training and used as the deep convolutional network model. It is particularly critical that the number of training times for each training of the deep convolutional network is monotonically positively correlated with the mean square error of pixel values ​​of the image where the current coding block is located. Content analysis is performed on the current coding block that is about to perform intra-frame coding to obtain various associated information of the current coding block, and the various associated information of the current coding block is the number of noise types, the maximum noise amplitude and the proportion of the number of edge pixels of the current coding block. At the same time, various associated information of a reference coding block of the current coding block is obtained, and the various associated information of the reference coding block The number of noise types, the maximum noise amplitude and the proportion of the number of edge pixels of the reference coding block, as well as the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, are obtained, thereby providing key basic information for the subsequent intelligent analysis of the amount of coding operations. The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various related information of the current coding block and various related information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model and obtain the amount of coding operations required for performing intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model. Therefore, before performing intra-frame coding, the different amounts of coding operations required for intra-frame coding of the current coding block using different reference coding blocks are intelligently analyzed, providing a reliable basis for the selection of reference coding blocks.

[0005] According to the present invention, there is provided a microcomputer algorithm operating system utilizing content parsing, the system comprising: An object reshaping device is provided in a microcomputer algorithm operating system and is used to parse a deep convolutional network model, in which the deep convolutional network is trained for each time to obtain a deep convolutional network after each training and used as the deep convolutional network model; A content analysis device is provided in a microcomputer algorithm operating system, and is used to perform content analysis on a current coding block that is about to perform intra-frame coding, so as to obtain various items of associated information of the current coding block, wherein the various items of associated information of the current coding block are the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the current coding block, and simultaneously obtain various items of associated information of a reference coding block of the current coding block, wherein the various items of associated information of the reference coding block are the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the reference coding block; A data extraction device, arranged in the microcomputer algorithm operating system, is used to obtain the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located; a parameter analysis mechanism, connected to the object reshaping device, the content analysis device and the data extraction device respectively, and used for inputting the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block into the deep convolutional network model, so as to run the deep convolutional network model and obtain the encoding operation amount consumed by the deep convolutional network model for performing intra-frame encoding on the current coding block using the reference coding block; The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the amount of coding operation consumed by using the reference coding block to perform intra-frame coding on the current coding block output by the deep convolutional network model includes: the length and width of the current coding block are respectively the same as the length and width of the reference coding block; The object reshaping device is arranged in a microcomputer algorithm operating system and is used to parse a deep convolutional network model, in which the deep convolutional network is trained for each time to obtain a deep convolutional network after each training and as the deep convolutional network model, including: the number of training times for each training of the deep convolutional network is monotonically positively correlated with the mean square error of pixel values ​​of the image where the current coding block is located; Among them, the data extraction device is set in the microcomputer algorithm operating system, and is used to obtain the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, including: the pixel value mean square error of the image where the current coding block is located is the mean square error of each pixel value corresponding to each pixel point of the image where the current coding block is located.

[0006] Therefore, the present invention has the following three outstanding technical effects: Technical effect A: an object reshaping device provided in a microcomputer algorithm operating system is introduced to analyze a targeted designed deep convolutional network model, in which the deep convolutional network is trained for each time to obtain a deep convolutional network after each training and used as the deep convolutional network model, and it is particularly critical that the number of training times of each training of the deep convolutional network is monotonically positively correlated with the mean square error of pixel values ​​of the image where the current coding block is located; Technical effect B: performing content parsing on a current coding block that is about to perform intra-frame coding to obtain various associated information of the current coding block, wherein the various associated information of the current coding block is the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the current coding block, and simultaneously obtaining various associated information of a reference coding block of the current coding block, wherein the various associated information of the reference coding block is the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the reference coding block, and obtaining the signal-to-noise ratio, pixel value mean square error, number of pixel rows, and number of pixel columns of an image where the current coding block is located, thereby providing various key basic information for subsequent intelligent analysis of the amount of coding operations; Technical effect C: The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns, length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block of the image where the current coding block is located are input into the deep convolutional network model to run the deep convolutional network model, and obtain the encoding operation amount required for performing intra-frame encoding on the current coding block using the reference coding block output by the deep convolutional network model, so that before performing intra-frame encoding, the different encoding operation amounts required for intra-frame encoding of the current coding block using different reference coding blocks are intelligently analyzed, providing a reliable basis for the selection of the reference coding block.

[0007] The microcomputer algorithm operating system using content analysis of the present invention is stable in operation and intelligent in design. By introducing an object reshaping device set in the microcomputer algorithm operating system, it is used to analyze the targeted designed deep convolutional network model, and then predict the amount of coding operations required to perform intra-frame coding on the current coding block using any reference coding block, thereby providing a reliable basis for the selection of the reference coding block. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein: Figure 1 The structure block diagram of the microcomputer algorithm operating system using content analysis is shown according to the first embodiment of the present invention. DETAILED DESCRIPTION

[0009] The following will describe in detail an embodiment of the microcomputer algorithm operating system utilizing content analysis of the present invention with reference to the accompanying drawings.

[0010] Figure 1 The structure block diagram of the microcomputer algorithm operating system using content analysis according to the first embodiment of the present invention is shown, and the system includes: An object reshaping device is provided in a microcomputer algorithm operating system and is used to parse a deep convolutional network model, in which the deep convolutional network is trained for each time to obtain a deep convolutional network after each training and used as the deep convolutional network model; For example, an object reshaping device is provided in a microcomputer algorithm operating system and is used to parse a deep convolutional network model, in which a deep convolutional network is trained for each time to obtain a deep convolutional network after each training and used as the deep convolutional network model, including: selecting and using a SOC chip to implement the object reshaping device, which is provided in a microcomputer algorithm operating system and is used to parse a deep convolutional network model, in which a deep convolutional network is trained for each time to obtain a deep convolutional network after each training and used as the deep convolutional network model; A content analysis device is provided in a microcomputer algorithm operating system, and is used to perform content analysis on a current coding block that is about to perform intra-frame coding, so as to obtain various items of associated information of the current coding block, wherein the various items of associated information of the current coding block are the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the current coding block, and simultaneously obtain various items of associated information of a reference coding block of the current coding block, wherein the various items of associated information of the reference coding block are the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the reference coding block; A data extraction device, arranged in the microcomputer algorithm operating system, is used to obtain the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located; a parameter analysis mechanism, connected to the object reshaping device, the content analysis device and the data extraction device respectively, and used for inputting the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block into the deep convolutional network model, so as to run the deep convolutional network model and obtain the encoding operation amount consumed by the deep convolutional network model for performing intra-frame encoding on the current coding block using the reference coding block; The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the amount of coding operation consumed by using the reference coding block to perform intra-frame coding on the current coding block output by the deep convolutional network model includes: the length and width of the current coding block are respectively the same as the length and width of the reference coding block; The object reshaping device is arranged in a microcomputer algorithm operating system and is used to parse a deep convolutional network model, in which the deep convolutional network is trained for each time to obtain a deep convolutional network after each training and as the deep convolutional network model, including: the number of training times for each training of the deep convolutional network is monotonically positively correlated with the mean square error of pixel values ​​of the image where the current coding block is located; The data extraction device is arranged in the microcomputer algorithm operating system, and is used to obtain the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, including: the pixel value mean square error of the image where the current coding block is located is the mean square error of each pixel value corresponding to each pixel point of the image where the current coding block is located; Among them, the various associated information of the current coding block are the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the current coding block, including: the ratio of the number of edge pixels of the current coding block is the ratio of the total number of edge pixels in the current coding block to the total number of pixels in the current coding block; The various items of associated information of the reference coding block are the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the reference coding block, including: the ratio of the number of edge pixels of the reference coding block is the ratio of the total number of edge pixels in the reference coding block to the total number of pixels in the reference coding block; Among them, the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block, and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the amount of coding operations required for using the reference coding block to perform intra-frame coding on the current coding block output by the deep convolutional network model also includes: performing binary value conversion on the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block, and various associated information of the reference coding block, and then synchronously inputting them into the deep convolutional network model; The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the reference clock signal used in the data processing process of obtaining the coding operation amount consumed by using the reference coding block to perform intra-frame coding on the current coding block output by the deep convolutional network model is generated by the signal generating element; The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model to obtain the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model. The reference clock signal used in the process is generated by a signal generating element, including: the signal generating element includes a quartz oscillation component; Among them, the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model to obtain the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model. The reference clock signal used in the signal generating element also includes: the quartz oscillation component generates a rectangular wave; The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model to obtain the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model. The reference clock signal used in the process is generated by the signal generating element, and also includes: using a programmable logic device to implement the quartz oscillation component; The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the reference clock signal used in the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model is generated by a signal generating element, and further includes: the programmable logic device is a GAL device; The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model to obtain the reference clock signal used in the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model, which is generated by the signal generating element and also includes: the programmable logic device is an FPGA device; And wherein, the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various related information of the current coding block and various related information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the reference clock signal used in the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model is generated by a signal generating element, and also includes: the programmable logic device is a CPLD device designed using VHDL language.

[0011] In addition, in the microcomputer algorithm operating system using content parsing, the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the encoding operation amount required for performing intra-frame encoding on the current coding block using the reference coding block output by the deep convolutional network model also includes: the encoding operation amount required for performing intra-frame encoding on the current coding block using the reference coding block output by the deep convolutional network model is in the form of binary numerical representation.

[0012] Although the invention has been described with reference to illustrative embodiments, it is not intended that this description be construed in a limiting sense. Various modifications and combinations of the illustrative embodiments, as well as other embodiments of the invention, will become apparent to persons skilled in the art upon reference to the description.

Claims

1. A microcomputer algorithm operating system using content analysis, characterized in that: The system comprises: An object reshaping device is provided in a microcomputer algorithm operating system and is used to parse a deep convolutional network model, in which the deep convolutional network is trained for each time to obtain a deep convolutional network after each training and used as the deep convolutional network model; A content analysis device is provided in a microcomputer algorithm operating system, and is used to perform content analysis on a current coding block that is about to perform intra-frame coding, so as to obtain various items of associated information of the current coding block, wherein the various items of associated information of the current coding block are the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the current coding block, and simultaneously obtain various items of associated information of a reference coding block of the current coding block, wherein the various items of associated information of the reference coding block are the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the reference coding block; A data extraction device, arranged in the microcomputer algorithm operating system, is used to obtain the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located; a parameter analysis mechanism, connected to the object reshaping device, the content analysis device and the data extraction device respectively, and used for inputting the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block into the deep convolutional network model, so as to run the deep convolutional network model and obtain the encoding operation amount consumed by the deep convolutional network model for performing intra-frame encoding on the current coding block using the reference coding block; The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the amount of coding operation consumed by using the reference coding block to perform intra-frame coding on the current coding block output by the deep convolutional network model includes: the length and width of the current coding block are respectively the same as the length and width of the reference coding block; The object reshaping device is arranged in a microcomputer algorithm operating system and is used to parse a deep convolutional network model, in which the deep convolutional network is trained for each time to obtain a deep convolutional network after each training and as the deep convolutional network model, including: the number of training times for each training of the deep convolutional network is monotonically positively correlated with the mean square error of pixel values ​​of the image where the current coding block is located; Among them, the data extraction device is set in the microcomputer algorithm operating system, and is used to obtain the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, including: the pixel value mean square error of the image where the current coding block is located is the mean square error of each pixel value corresponding to each pixel point of the image where the current coding block is located.

2. The microcomputer algorithm operating system using content analysis as claimed in claim 1, characterized in that: The various associated information of the current coding block are the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the current coding block, including: the ratio of the number of edge pixels of the current coding block is the ratio of the total number of edge pixels in the current coding block to the total number of pixels in the current coding block; The various items of associated information of the reference coding block are the number of noise types, the maximum noise amplitude, and the ratio of the number of edge pixels of the reference coding block, including: the ratio of the number of edge pixels of the reference coding block is the ratio of the total number of edge pixels in the reference coding block to the total number of pixels in the reference coding block; Among them, the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various related information of the current coding block and various related information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the encoding operation amount required for using the reference coding block to perform intra-frame encoding on the current coding block output by the deep convolutional network model also includes: the signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various related information of the current coding block and various related information of the reference coding block are converted into binary values ​​respectively and then synchronously input into the deep convolutional network model.

3. The microcomputer algorithm operating system using content analysis as claimed in claim 2, characterized in that: The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the reference clock signal used in the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model is generated by the signal generating element.

4. The microcomputer algorithm operating system using content analysis as claimed in claim 3, characterized in that: The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various related information of the current coding block and various related information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model to obtain the reference clock signal used in the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model, which is generated by a signal generating element, including: the signal generating element includes a quartz oscillation component.

5. The microcomputer algorithm operating system using content analysis as claimed in claim 4, characterized in that: The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various related information of the current coding block and various related information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model to obtain the reference clock signal used in the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model. The signal generating element also includes: the quartz oscillation component generates a rectangular wave.

6. The microcomputer algorithm operating system using content analysis as claimed in claim 5, characterized in that: The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various related information of the current coding block and various related information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model to obtain the data processing process output by the deep convolutional network model, which uses the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block to perform intra-frame coding on the current coding block. The reference clock signal used in the process is generated by a signal generating element and also includes: using a programmable logic device to implement the quartz oscillation component.

7. The microcomputer algorithm operating system using content analysis as claimed in claim 6, characterized in that: The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various related information of the current coding block and various related information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the reference clock signal used in the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model is generated by a signal generating element, and also includes: the programmable logic device is a GAL device.

8. The microcomputer algorithm operating system using content analysis as claimed in claim 6, characterized in that: The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various related information of the current coding block and various related information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model, and the reference clock signal used in the data processing process of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block output by the deep convolutional network model is generated by a signal generating element, and also includes: the programmable logic device is an FPGA device.

9. The microcomputer algorithm operating system using content analysis as claimed in claim 6, characterized in that: The signal-to-noise ratio, pixel value mean square error, number of pixel rows and number of pixel columns of the image where the current coding block is located, the length and width of the current coding block, various associated information of the current coding block and various associated information of the reference coding block are input into the deep convolutional network model to run the deep convolutional network model to obtain the data processing process output by the deep convolutional network model of the coding operation amount required to perform intra-frame coding on the current coding block using the reference coding block. The reference clock signal used in the process is generated by a signal generating element and also includes: the programmable logic device is a CPLD device designed in VHDL language.

Images