Research On Enhancement Algorithm Of Foggy Image

In recent years,the haze weather has been frequent in China,so the image quality of video monitoring system has been reduced and the recognition of image is difficult.Therefore,it is of great practical significance to study the aerosol image enhancement algorithm.In view of the problems of homomorphic filtering algorithm and histogram equalization algorithm,this paper presents a new algorithm for the combination of the two.Correction using improved homomorphic filtering algorithm of uneven illumination of the image brightness,nonlinear problem with logarithmic linearization,and the low frequency and high frequency component separation,use the high-pass filter suppress low frequency information,and histogram equalization,the low frequency components through non-linear stretching the pixel values of the region.The algorithm enhances image contrast,enhances detail and profile information,and makes foggy images more visible.However,it is not obvious that the image can be used in the distance after processing.Aiming at the shortcomings of the above algorithm and the high time complexity of fog algorithm,the improved dark channel and MSR foggy image enhancement algorithm are proposed.First with two guide filter combined with the minimum filter method,optimum transmissivity of dark grey prior model,and then in the HSV color space,the bilateral filter function is adopted to estimate the brightness of the image V,the brightness of the incoming component using gamma curve correction,the S type curve reflection components of image contrast stretching processing,then the output image in RGB space.The improved algorithm can effectively restore the color of foggy image,the edge information is more complete,the halo phenomenon is improved,and the visibility of the image is improved.However,the efficiency of the algorithm is relatively low,and it is not very good for fog weather image.In view of the serious problems of fog weather image degradation,an improved morphological and MSR foggy image enhancement algorithm is proposed.Firstly,the image is filtered by multi-scale morphology based on top hat,and the image noise is suppressed by the large scale structure element,and the edge information of the image is maintained by the small scale structure element.And then use the improved MSR algorithm to enhance the brightness V,using direct filtering instead of gaussian filtering function to estimate of light weight,keep on the edge of the image at the same time improve the operation rate of the algorithm,finally the image converted to RGB space.Improved algorithm to fog effect,the noise and the halo phenomenon are suppressed,the image is clearer,the color is more natural.This method is very effective for fog image enhancement and is not suitable for the treatment of mist image.