基于梯度变换的多传感器图像融合算法  

Multi-Sensor Image Fusion Algorithm Based on Gradient Transform

作　　者：王建[1] 刘肖[1] 王国珲[2] 

机构地区：[1]西安航空学院计算机工程系,西安710077 [2]西安工业大学光电工程学院,西安710032

出　　处：《重庆理工大学学报（自然科学）》2012年第10期62-65,共4页Journal of Chongqing University of Technology：Natural Science

基　　金：陕西省教育厅科研计划资助项目(11JK0996)

摘　　要：针对像素级多传感器图像的融合问题,设计了一种基于梯度变换的多传感器图像融合算法。该算法利用金字塔变换得到图像的多分辨序列,在相应层次的金字塔图像上进行多方向分解,采用基于梯度变换的融合方法以获取最终的图像。从目视效果来看,采用该算法进行图像融合,能将2幅多传感器图像融合成1幅信息量饱满的单一图像。实验结果表明,该算法与同类其他算法相比较,标准差提高了0.05%～13.66%,偏差度降低了10.33%～48.49%,联合熵提高了0.36%～7.22%。Focusing on the fusion problem of the multi-sensor images for pixel level, the multi-focus image fusion algorithm based on regional energy was designed. By using muhiresolution image sequences obtained from the pyramid transform, the source images were decomposed on the corresponding level of multi-direction, and the source images were fused on each corresponding levels with the gradient transform method, then the fused image was achieved through inverse pyramid transform. From the visual effect, the algorithm can fuse into a full information content image from two-sensor im- age. Experimental results and comparisons with three traditional algorithms demonstrate that the designed algorithm is superior with 0.05% to 13.66% improvement of standard deviation, 10.33% to 48.49% reduction in the deviation degree and 0.36% to 7.22% enhancement in union entropy.

关 键 词：多传感器图像 梯度变换 图像融合 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]