Reach On Technology Of Video Clarity In Weather Of Rainy

To obtain information from the outside world, the information accounted for more than70%from visual. Image information has increasingly become the important way of human to obtain information. With the rapid development of image processing and computer vision technology, more and more computer vision system used in many field of science and engineering. However, the bad weather makes the visual effect and data quality of image has fallen sharply and brings enormous difficulties for detection and recognition in vision system.In recent years, image processing research of bad weather has made great progress, but for a video in weather of rainy real-time processing still lack of systematic and comprehensive analysis. In present, there are two main types of rain removal of video. One is frame difference method, the other one is K-means method. But the two methods are deficiencies in processing effect and processing speed. A processing method of image clarity in weather of rainy was proposed, which has certain advantages in processing effects and processing speed.The main work and results involve:Firstly, through the analysis of the raindrops physical properties and optical properties, combined with the rain test and get rid of the time requirements of real-time processing, four-frames difference method was proposed to detect the rain in current frame.Secondly, to improve four-frames difference method. On the basis of the same in the treatment effect, rain field removal method is improved by reducing the number of detection indicators in order to improve the processing speed of the algorithm. And based on the raindrop detection method, we proposed two-step method to remove raindrops of current frame.Thirdly, the difference detection system of raindrop is built and related test program is carried out by utilizing MATLAB program. Simulation experiment proves that this method has high processing speed. The feasibility of the method has been verified by experiment.Finally, a detection index which B component act a quantitative effect of rain removal was proposed by analyzing the recovery image clarity and image information entropy.