Single Image Fog Removal Algorithm Based On Non-local Information

As countries pay more and more attention to the development of science and technology,people's requirements for image quality are still gradually increasing.Sound and pictures are the contents that human beings can directly exchange in the daily life.The images stand in a pivotal position in the study of vision,because it can directly express the communication to the outside world.Many application systems,such as security monitoring systems,military surveillance systems,highway monitoring systems,and remote sensing navigation systems,require accurate detail analysis and image feature extraction.However,in reality,there is no way to avoid foggy bad weather.This will result in images that are captured in this case with poor detail and damage in terms of usability.The quality of picture scenes obtained by many imaging systems is degraded,and the contrast and sharpness of the picture will be affected.Therefore,in order to improve the usability of the image,it is necessary to process the image to obtain the satisfactory result we want.Based on the analysis of the background and concept of fuzzy foggy image quality,we tried and improved the dark channel prior defogging algorithm.The main tasks of this paper are as follows:(1)based on the atmospheric physical model,the block effect and distortion in the dark channel fog removal algorithm are improved.Based on the dark channel defogging algorithm,the estimated value of dark channel in the boundary area of sky and non-sky is relatively small,and the block effect is easy to occur.In view of the problems in the process of defogging,the multi-scale method is used to eliminate the high light,so that the distortion and blocky effect in the process of image defogging are improved.(2)further improve the image defogging algorithm of He,and use the fuzzy line method to further obtain the function relation of atmospheric light,so as to refine the atmospheric light value.Furthermore,the transmittance in the physical model is further calculated by the refined atmospheric light value.The experimental results were compared and analyzed.The experimental results show that the improved image has obvious defogging effect and the comprehensive value is greatly improved.