Research On Video Image Fogging System In Smog Environment

With the development of technology,people no longer simply rely on their own sensory systems to obtain outside information.In order to meet the needs of people to obtain external information,the development of machine vision equipment is very rapid,but in the case of haze in outdoor scenes,video images acquired through machine vision equipment often have low contrast,and even lead to some key information.Lost.This makes computer applications such as road monitoring and car vision not working properly.In order to reproduce the key information of the foggy video image obtained,this paper optimizes the video defogging technology based on the single image defogging,and finally obtains a clear and smooth video sequence.Based on domestic and international research,the current technology for static foggy image processing is more mature than video fogging technology.Since most of the current algorithms for video dehazing are based on single-frame image processing,this paper first analyzes and studies the atmospheric imaging model and the dark primary color prior algorithm in the presence of haze,and then The channel prior algorithm solves the problem of sky region failure.This paper proposes an improved tolerance mechanism algorithm to estimate the transmittance and correct the transmittance.At the same time,the guiding matrix is reconstructed to optimize the initial transmittance,so as to achieve fast defogging.At the same time,the dark channel image is divided into sub-blocks of size 4*3,and then the sub-blocks are sorted by the method based on distributed sorting to re-find the position of the atmospheric light value.In addition,based on the single-frame image dehazing algorithm,this paper optimizes the solution of ambient light value and transmittance value and the selection of parameters in the video sequence.In order to verify the real-time and stability of the video dehazing algorithm,the video defogging algorithm is simulated on the Matlab platform.According to the video results of the simulation output,the recovered video sequence has no flicker and jamming,which proves that the proposed algorithm has good robustness and real-time performance.Finally,the software simulation algorithm is written and designed by hardware programming language(Verilog HDL),and finally mapped to the FPGA platform.According to the FPGA image development process,the outdoor video image with FPGA as the hardware platform is built and realized.The fog system enables video defogging through a hardware system.Through the online system simulation and comprehensive debugging,the fog-through system designed in this paper has high real-time and stability.