| In recent years,smoggy weather has become a social issue.Fog weather have an impact on human health and many other aspects of human society.In some severe weather condition,the pictures or videos captured by monitoring equipments may have poor quality due to the scattering process of the paticles floating in the air.These image degradation includes image blurring and low contrast,which greatly effected the operator to observe.Therefore,image defogging is particularly necessary.In this paper,we first study the defog algorithm of single image,summarize the research results of predecessors in the field of image defog,and put forward the concept of foggy coefficient on the basis of Dark channel prior,which can reflect foggy degree of the current scene At the same time,the CLAHE algorithm and the histogram equalization algorithm are optimized and improved.Combining the foggy coefficient and the improved algorithm,a single image defog algorithm based on Dark channel is proposed.The algorithm not only considers the physical model of fog image degradation,but also uses the image enhancement theory based on histogram processing.Based on the study of single image defog algorithm,this paper makes several optimization and improvement of the algorithm,and further improves the efficiency of the algorithm while ensuring the excellent defog effect.Then,a video image fog algorithm based on Dark channel has been proposed.Finally,KC705 development kit provided by Xilinx has been used as the core chip to build a video image defog system,and through the system in the FPGA platform to achieve this dark channel based video image defog algorithm.The results show that the algorithm can effectively achieve image defogging,while eliminates stains,and remains the real color of the image.The image and video obtained by the algorithm has high contrast and visibility,which means good image quality.Meanwhile,the efficiency is higher,and the algorithm can effectively deal with 60 fps foggy video. |
PDF Full Text Request