Research Of Haze Removal Algorithm And Implement Based On Qt Platform

In recent years,with the degradation of air quality more and more serious,haze and other bad weather appear more and more frequently.Due to the haze and other adverse weather conditions,computer and other image acquisition system to receive the scene of the object of the reflected light process,by the atmospheric environment of the scattering of particles in the scattering and absorption of the role of the outdoor image or video information to be disturbed,Causing the details of the image blurred,dark colors,poor visibility,resulting in the haze in the weather under the image or video,its quality is seriously degraded.These degraded images not only make the utilization value of the system lower,narrow the scope of application,but also greatly hinder the computer image acquisition system to extract the image features,and seriously affected the subsequent image processing and analysis.Therefore,it is of great theoretical and practical significance to study the fast and effective clarity of fog images.In this paper,a series of studies on the clarification of fog images are carried out from two aspects: image restoration and image enhancement.We focus on the image de-fog algorithm based on the prior theory of dark channels and histogram equalization image enhancement algorithm.To achieve a fast and effective image to the fog algorithm,the specific research as follows:1)The image defogging method based on atmospheric scattering model is studied deeply.Based on the atmospheric scattering model,the image de-fog algorithm based on the prior theory of dark channels is studied emphatically.The performance of the algorithm is verified by simulation experiments,and the shortcomings of the algorithm are analyzed.Then,based on the shortcomings of the algorithm and the inability to meet the real-time requirements,the paper proposes a new method to improve the transmittance and atmospheric light by using the mean filter.The experimental results show that the improved algorithm is fast and fog good results.2)For the image restoration method based on the physical model,the brightness and color of the image are dim after the de-fogging process,and the image enhancement method is proposed by using the histogram equalization(HE)algorithm.Histogram equalization is a commonly used image contrast enhancement technique,but HE can not satisfy the processing of local information.Local histogram equalization(LAHE)algorithm has made some improvements to the HE,but will produce a more obvious block effect,based on the use of a restricted contrast local histogram equalization(CLAHE)algorithm,by limiting the contrast control noise excessive amplification,But also can effectively highlight the details of the image information.Subsequently,the speed is calculated by interpolation.3)Based on the Qt development platform,the improved dark channel algorithm is used to de-fog the image,and then the contrast histogram equalization algorithm is applied to the image enhancement after de-fog.By combining these two algorithms,The image is reconstructed and the image enhancement operation is carried out.A single image de-fogging software is developed to realize the de-fogging function of the algorithm,and the parameters are adjustable and the de-fog effect is good and the real-time demand is satisfied.