Image Haze Removal Improved Algorithm Using Dark Channel Prior

In rece nt years,environmental pollution is becoming more and more serious.Because of haze,the visibility is greatly reduced,making it impossible for people to get accurate information.So people tend to make a wrong judgment which sometimes can be so serious as to cause a disaster.When there is fog,due to a large number of suspended particles floating in the air,the intensity of the light will be reduced because it is scattered in the process of transmission.The attenuation of the light intensity reduces the image contrast,the details,and the color fidelity.Dr.He Kaiming's Dark channel fog removal algorithm is a major breakthrough in the field of image defogging Technology.The algorithm is simple but effective,real-time and automatic defogging,while the problem is its shortcomings are mainly on the atmospheric scattering physical model over-reliance,can not choose to match the size of the filter template,the lack of applicability of the sky and the image after the dark side.Based on the atmospheric scattering model,this paper analyzes the causes of the decline of image quality in fog,and experiments,improves and tries the dark channel de-fog algorithm,and achieves good imaging results.The main work of this paper can be summarized as follows:(1)A method based on multi-scale is proposed to refine the transmittance so as to achieve the effect of defogging.Based on dark channel prior criterion,this method makes sure that the dark channel prior criterion and hypothesis of the same transmittance in the region stand and at the same time,prevents the occurrence of block effect of transmittance image in the region of dark channel mutation.(2)The adaptive filter is used to enhance the contrast and maintain the structural information when the dark channel is obtained.The process of obtaining the original atmospheric light value is improved,which can effectively restrain the phenomenon of excessive color saturation after defogging of the sky.(3)For transmission estimation,in order to overcome the sky area after showing halo color restoration,this paper achieves good restoration effect by analyzing the transmission characteristics,providing special treatment in the middle of the process before the initial transmission,then guiding the filtering optimization.The results of the experiments fully show the good effect of the proposed method,and compared with other methods,it is simple and more effective.However,this method is not suitable for some kinds of fog image,especially images contain a large areas of sky and brightness towards which the effect of this method is not ideal,and it is worthy of further study and improvement.