A Video Fractional Differential Operator Enhancement System

Abstract

The invention discloses a video fractional differential operator enhancement system, comprising a power supply module for power supply of the whole machine, and a video enhancement system based on a ZYNQ platform electrically connected with the power supply module; the video enhancement system based on the ZYNQ platform comprises: A ZYNQ processor for reading the decoded video stream, partitioning and buffering the video stream, and extracting and processing the video images in the buffer area; and connecting with the input port of the ZYNQ processor and used to convert the encoded digital video A video decoding chip for restoring and decoding the signal; and a DDR3 SDRAM memory that is connected to the ZYNQ processor and used to partition and cache the video stream; and a video that is connected to the output port of the ZYNQ processor and used to output the enhanced video stream. An output module; the video fractional differential operator enhancement system and its implementation method of the present invention realize real-time video enhancement processing.

Description

technical field [0001] The invention relates to a video fractional differential operator enhancement system, which belongs to the technical field of video image processing. Background technique [0002] Fractional-order calculus operation is a generalization of integer-order calculus operation. Compared with integer-order calculus operation, fractional-order calculus expands the order of operations. There is a higher degree of freedom and flexibility in implementation; most of the signals in modern signal analysis and processing are unstable signals with various characteristics, these signals have nonlinear, non-causal, non-stationary and other characteristics, so in the theory of fractional calculus, the Various algorithms are available for processing such signals. In recent years, the research and application of fractional calculus in the field of signal analysis and processing has made rapid progress, and the combination of image signal as a two-dimensional signal and fr...

AI Technical Summary

Problems solved by technology

Fractional calculus has also made rapid progress in the field of image processing, which is mainly reflected in the application of image detail enhancement, image edge extraction and image denoising processing; the wide application of fractional calculus in the field of image processing mainly benefits from Based on the advantages of the weak derivative of fractional calculus, it can enhance the high-frequency components of the image while retaining the low-frequency components of the image non-linearly; video enhancement processing is to enhance the received video stream signal, such as the remote transmission Low-bit-rate video is converted into high-definition and clear pictures. In the existing technology, the video content is not clear and the video is compressed to ensure that the noise can be viewed at the receiving end. AI technology intelligently removes artifacts, denoises and other technologies to remove the video Compression artifacts and noise in the image, thereby improving the picture. Among them, there are many video enhancement processing methods, such as Chinese patent application number: 201811198242.0, which discloses a video enhancement method. The method includes: obtaining the resolution of the video to be displayed; obtaining The video enhancement algorithm corresponding to the resolution; through the video enhancement algorithm, the video to be displayed is enhanced; the program selects the video enhancement algorithm according to the resolution of the video, and the enhancement of the video is more in line with the characteristics of the video itself, so as to obtain more Good video enhancement effect, but it is correspondingly enhanced for the video resolution, and the video stream cannot be enhanced. Another example is the video enhancement method disclosed in Chinese patent application number: 201410401578.8, which sequentially acquires each frame in the video; For each current frame, determine the attribute information reflecting the motion state or light and shade, search the preset mapping relationship to obtain the control parameters corresponding to the attribute information, and use the control parameters to perform video enhancement operations on the current frame; the operation is By controlling the attribute parameters of each frame, the computational workload is large, and it cannot be applied to the enhancement of video data of PDAs and other portable devices. However, the existing CPU-based video enhancement processing has poor real-time performance and is generally oriented towards image enhancement rather than video. Enhancement; the enhancement algorithm with good effect is difficult to be extended to the field of real-time video enhancement due to insufficient processing hardware computing power; fractional differential operator has attracted extensive attention because of its good enhancement effect, but because its algorithm is more complicated than conventional algorithms, it is difficult to Real-time video enhancement on CPU-based platform

Images