结合K-SVD算法与Canny边缘检测的DTCWT域图像融合方法

Image Fusion Method of the K-SVD Algorithm and Canny Edge Detection in the DTCWT Domain

为了快速有效的获取高质量的融合图像,利用具有线性以及多方向选择性的双树复小波(DTCWT)对源图像进行分解,结合K-SVD(K奇异值分解)算法获取训练字典进行低频信号的稀疏融合,同时将Canny算子与区域能量法结合进行高频信号的融合。选用两组红外与可见光图像进行融合实验,对比传统的小波与DTCWT融合方法,获得了更清楚、更准确的高质量图像,同时保留了图像更多的边缘信息。

To obtain high quality fusion images quickly and effectively,the image is decomposed by dual-tree complex wavelet transform(DTCWT)with good linear characteristics and direction selectivity.The training dictionary is acquired by combining K-SVD(K singular value decomposition)algorithm for sparse fusion of low frequency signals,and the Canny operator is combined with region energy method for high frequency signal fusion.By using two groups of infrared and visible images for fusion experiments,compared with the traditional wavelet and DTCWT fusion methods,a clearer and more accurate high-quality image is obtained,while retaining more edge information of the image.