摘要
目前,雾霾所引起的图像模糊问题,主流的算法主要都侧重于处理雾气,对于霾没有相关的处理.针对此缺陷,提出了一种联合K-SVD(K-singular value decomposition)稀疏算法和暗通道先验算法的全新算法,来克服雾霾引起的图像模糊问题.图像的处理主要分两个步骤:第一步是运用KSVD稀疏算法去除图像中的霾恢复出只含雾气的图像,第二步通过经典的暗通道算法去除图像上的层层雾气.计算机仿真结果表明,该方法对于图像的处理结果要优于FVR(Fast visiblity restoration)算法,暗通道先验算法和直方图均衡化算法.
At present, to solve the problem of image blur caused by haze, most algorithms focus on eliminating blur caused by mist but ignore the blur caused by suspended particles. Aiming at solving the problem mentioned above, a novel algorithm based on K-SVD sparse algorithm and the dark channel priori algorithm is proposed. The algorithm can be divided into two steps： first, K-SVD sparse algorithm is employed to remove the dust particles in the image, obtaining a recovered, fog-only image. Second, the fog in the image is removed by means of the classic dark channel algorithm. The simulation results illustrate that the proposed algorithm is better than FVR algorithm, dark channel priori algorithm and histogram equalization algorithm, which proves the superiority of this method.
出处
《浙江工业大学学报》
CAS
北大核心
2017年第3期315-319,共5页
Journal of Zhejiang University of Technology
基金
国家自然科学基金资助项目(U1509219
61471322
61402416)
