| In the foggy weather environment,the contrast of the image captured by the highway monitoring system is low,and the target in the image is fuzzy,which makes it difficult to monitor the driving vehicles.Therefore,image defogging is of great significance in highway monitoring.The purpose of image defogging is to improve image contrast,make the target in the image clear,and make monitoring system can accurately extract important information of driving vehicles,mainly including the details of vehicle model and license plate number.Therefore,image defogging algorithm is an important research direction.Image defogging models are now mainly divided into two categories: fogging image enhancement models and dark channel defogging models.These two models can improve the clarity of the image,so as to achieve the effect of defog.In this paper,we focus on the haze-based prior single image dehazing algorithm.This algorithm can obtain good results for some foggy images,but has the disadvantage of inaccurate prior information,and cannot meet the real-time requirements well.In view of this paper,improvement algorithms are proposed.Improved algorithm 1: Firstly,the image is pre-processed by using the guide map filtering method,and then the obtained filtered image is processed by the dark channel defogging algorithm.The improved algorithm 2: Firstly,using the information of the fog-free image as the prior information of the dark channel to replace the dark channel information in the original algorithm,and then the dark channel defogging algorithm is used to defogging the foggy image.In order to further illustrate the effectiveness of the algorithm in this paper.In the experimental part of this paper,entropy,peak signal to noise ratio,edge gradient ratio and average gradient were used to verify that the improved algorithm effectively improved the image clarity and image detail integrity,indicating that the improved algorithm is better than the traditional dark channel algorithm. |
PDF Full Text Request