Research On Image Dehazing Algorithm Based On Dark Channel

Haze affects every aspect of daily life,such as travel problems,health problems,and so on.The rapid development of artificial intelligence technology has enabled many human behaviors to be replaced by machines,allowing humans to explore more unknown fields.Among them,computer vision is a particularly important component of artificial intelligence technology.The clarity of the input image is an indispensable prerequisite for various computer vision technologies such as recognition and tracking.However,outdoor images in haze weather are often blurred,and it is difficult to obtain effective information in the image.Therefore,it is particularly important to dehaze the image.This paper studies the popular image defogging algorithms at home and abroad.Image defogging algorithms can be divided into two aspects according to different focuses.One is the defogging algorithm based on image enhancement,and the other is defogging based on image recovery.algorithm.This paper selects the defogging algorithm based on the dark channel prior to study,introduces and analyzes the overall process in detail,and modifies the transmittance optimization method for the efficiency of the algorithm.The dark channel prior theory is not applicable to the sky area in the foggy image,which causes the sky area in the foggy image to be distorted seriously and generate a lot of noise at the same time.In view of the above problems,this paper proposes a dark channel defogging algorithm based on the extraction of the sky region.First,the edge information of a single input image is sharpened by the Retinex algorithm,and the enhanced image is extracted by the edges through the Canny operator,and the sky area is accurately extracted by combining Close;second,the atmospheric light A in the sky area is more accurately selected.In order to solve the problem of color distortion in the sky area,the sky area in the image is converted from the RGB color model to the HSI color model for processing,and the brightness channel I is processed by the improved dark channel dehazing algorithm,which retains the hue information and The problem of saturation information and additional noise is solved through a new transmission optimization method.Finally,the non-sky area is processed by the improved dark channel defogging algorithm to complete the defogging of the entire image.The defogging algorithm proposed in this paper and a variety of classic defogging algorithms are simulated.Through comparison and analysis of the simulation results,it is confirmed that the proposed method well solves the problems of color distortion and noise increase in the sky area,and makes the image after defogging more effective Clear and natural.