Research On Image Defogging Method Based On àTrous Algorithm

The fog image is obtained by taking pictures of outdoor electronic equipment in bad weather.This kind of image can be obtained in all foggy places.However,in the process of obtaining,it will be caused by fine particles such as fine sand in the air.The quality of the obtained images is very low,most of them have problems such as blurry details,low contrast,gray and white colors.Xinjiang’s unique geographical environment makes the fog images in Xinjiang unique.Not only these common problems exist,there is also a lot of dust in the image,which affects our analysis and understanding of the image of fog in Xinjiang.Therefore,the defogging processing of Xinjiang fog images is particularly important.An effective defogging method can not only remove the fog in the image,but also cover the dust in the image.While the image content is highlighted,it can restore the color of Xinjiang fog image and reduce the fuzziness of Xinjiang fog image.Based on the current image defogging method,we will delve into the basic theory and lifting scheme of Dyadic wavelet transform and its improvement schemes as well as the properties and applications of traditional defogging methods.Two new methods for image defogging are proposed.Research the related theories of Dyadic wavelet transform,and construct a new B-spline dyadic wavelet filter,and propose a new method of image defogging combining the Dyadic wavelet method and the gray world method.The two-dimensional àTrous algorithm is used to decompose and reconstruct the Xinjiang fog image in the input Dyadic wavelet domain with different filters,and the low-frequency sub-band coefficients and the high-frequency sub-band coefficients use different functions for defogging processing;Using the traditional defogging method to improve the overall contrast of the reconstructed fog image,and combine the gray world method for color correction.The final experimental results show that the two filters have different defogging effects under the same method and process,and the defogging processing of the Xinjiang fog image by the method in this chapter can subjectively make the content of the image covered by sand and dust prominent.A new method of image defogging based on Dyadic wavelet method combined with homomorphic filtering method is proposed.Through research and analysis of the shortcomings of traditional defogging methods,the traditional defogging methods are improved,and combined with the àTrous algorithm,a new method for defogging Xinjiang fog images is obtained.Compared with the classic defogging method,this method has a certain effect on the defogging processing of Xinjiang fog images.