Research On Single Image Dehazing Based On Improved Dark Channel Prior Algorithm

Image dehazing is a technique for clearing the image of the fogged image obtained in the haze weather through the dehazing algorithm.The dehazing algorithm can be used to obtain high-quality restored images.In recent years,haze weather often occurs with the development of industry and the deepening of urbanization.Image degradation in haze weather is not conducive to subsequent operations such as image feature extraction,making computer vision systems unable to function well and bringing a lot of inconvenience to life.Therefore,real-time and effective dehazing of fogged images has important research value and significance.In this paper,the traditional Dark Channel Prior(DCP)dehazing algorithm has been deeply studied.Aiming at the problems and shortcomings in the traditional algorithm,this paper proposes the corresponding improved algorithm.The main work of this paper is as follows:(1)The traditional DCP algorithm has the problems of poor robustness for single window and high time complexity of image matting.To solve this problem,this paper proposes an algorithm named Image Defogging Algorithm Based on Guided Filtering and Optimized Atmospheric Light Value(IDAGA).Based on atmospheric scattering model,the method refines the transmittance of atomized images by fitting windows of different sizes and guided filtering.At the same time,the combination of open operation in image morphology and bilateral filtering are adopted to reduce the noise of white objects on the atmospheric light value and obtain more accurate atmospheric light value.Finally,color compensation is conducted based on HSV color space to solve the problem of low brightness of restored image.(2)Since the dark channel algorithm is not applicable to bright areas such as the sky,the restored image will be distorted in such areas.Besides,the restored images will have lower brightness and contrast in the traditional dark channel algorithm.In order to solve this problem,a new dehazing algorithm named Image Defogging Algorithm Based on Adaptive Tolerance and CLAHE(IDATC)is proposed.The method uses adaptive tolerance to correct the transmittance of high brightness areas such as the sky,solving the problem of distortion and ensuring the image quality in such areas after fog removal,so as to obtain high quality restored images.At the same time,the L component in the Lab color space of the image is processed by CLAHE algorithm to achieve the enhancement of the restored image.(3)The evaluation criteria for image dehazing are constructed from subjective evaluation and objective evaluation.Comparative experiments are designed to verify the effectiveness of the algorithm proposed in this paper by comparing the two algorithms,IDAGA and IDATC,with other dehazing algorithms.