Research On Image Defogging Algorithm Based On Atmospheric Scattering Model

Since entering the 21st century,with the development of domestic industry,the national economy has grown significantly,modern life has become more and more convenient,and the national happiness index has been greatly improved.However,at the same time,environmental pollution has become more and more serious.The ensuing smog phenomenon occurs frequently in most areas of China,especially in urban areas with high population density.Haze affects people's respiratory system and can cause diseases such as acute rhinitis and acute bronchitis;and it can also lead to the weakening of ultraviolet light near the surface,thereby increasing the activity of infectious pathogens in the air and leading to an increase in infectious diseases;It will seriously affect the normal operation of society.The technical tools related to national economy and people's livelihood,such as city monitoring,autonomous navigation,remote sensing detection,military reconnaissance,and aerial photography,all rely on collecting clear images for analysis and ensuring the normal operation of human society.Therefore,under foggy weather conditions,it is particularly important to perform further defogging processing on the collected foggy image to obtain a clear image.At present,there are various methods for image defogging.Among them,the image defogging method using the atmospheric scattering model is different from other traditional image defogging methods.The atmospheric scattering model is a good interpretation of light(including reflection,refraction,and scattering)in the atmosphere.The effect of the transmission is more realistic.However,the current image defogging method based on the atmospheric scattering model also has certain shortcomings.On the one hand,there is a "halo" phenomenon(halo)in the area where the image depth of field changes suddenly(the edge of the scene in the image),and the other On the one hand,there is severe color distortion after the sky area in the image is restored.In view of the above shortcomings,this thesis studies and improves the image defogging algorithm based on atmospheric scattering model.The main contents of this article are as follows:(1)This thesis proposes an edge-weighted guided filtering algorithm.The edge value of the transmittance obtained after the guide filtering process for the transmittance in the atmospheric scattering model is not accurate enough,which leads to the occurrence of halo phenomenon at the edge of the scene in the restored image.In this thesis,by using guided filtering combined with the edge detection capability of Canny edge operator to adaptively weight the guided filtering image,while ensuring the speed of the guided filtering process,the edge retention of the filtered image is improved.The experimental results show that the edge-weighted guided filtering algorithm proposed in this thesis has high efficiency,and better protects the edges of the image and avoids the occurrence of halos.(2)This thesis proposes an image defogging algorithm with improved tolerance mechanism.Aiming at the dark channel prior theory used in the image defogging algorithm based on the atmospheric scattering model,the dark channel prior theory fails in some areas of the image(such as the sky,white cars,and other areas with high brightness).In this thesis,by adding an improved tolerance mechanism to the atmospheric scattering model,the compensation in the part that does not meet the dark channel prior theory area tends to be smooth,effectively improving the color distortion phenomenon in the original method.The experimental results also show that adding an improved tolerance mechanism to the atmospheric scattering model restores the fullness and realism of the image,and avoids the phenomenon of color distortion.(3)This thesis combines the above algorithm to design and implement an image defogging system based on Android platform.The system uses an Android phone to take a foggy image or gallery to take a picture in a real environment,and then transfers the image to the background server,which performs the image defogging process.Finally,the background returns a clear image to the Android phone and saves it on the phone in the album.The operation test of the system shows that the method proposed in this thesis is practical and effective,and shows the feasibility of the application of the image defogging algorithm in the industry.