基于NSCT的多源图像融合规则研究  

Research of multi-source image fusion rules based on NSCT

作　　者：罗杰 孔韦韦[1] 

机构地区：[1]武警工程大学,西安710086

出　　处：《激光杂志》2016年第11期132-135,共4页Laser Journal

基　　金：青年科学基金项目(61309008;61309022)

摘　　要：图像融合作为图像处理领域的重要组成部分之一,在医疗诊断、军事侦查、遥感测绘等诸多方面有着广泛的应用。结合当前图像融合领域内比较热门的非下采样轮廓波变换(Non-subsampled Contourlet Transform,NSCT)方法,根据图像融合的流程,对变换后高、低频分量系数的融合规则加以总结。通过调研引用量较大、成果较突出和融合实验效果显著的相关文献采用的融合规则,将多源图像的融合规则分为基于像素、基于区域、基于视觉特性和基于矩阵分析四类,并对其特点和应用做了系统的归纳,最后对图像融合规则未来的发展做出了展望。As one significant components of the field of Image Processing, muhi-source image fusion is widely employed in many aspects, such as medical diagnose,military reconnaissance and remote sensing survey. The paper sums up the fusion rules of the high and low frequency coefficients in the domain based on the NSCT （ Non-subsampled Contourlet Transform）which is popular in the field of Image Fusion at present. The relevant literature which are higher cited, more prominent and remarkable in the fusion experiment are analysed. The fusion rules could be divided into four parts： pixel, visual characteristic and matrix analysis. Then, the features and application of the several rules are systematically induced and concluded. In the end, the paper look forward to the future of the development of image fusion rules.

关 键 词：图像融合 NSCT 融合规则 像素 区域 

分 类 号：TN391[电子电信—物理电子学]