摘要
为了把图像的融合效果评价信息加入到融合规则的选取过程当中,针对合成孔径雷达和可见光两种常见的遥感图像,提出了一种基于区域性能的自适应融合方法。该方法首先利用平滑滤波抑制SAR图像斑点噪声,再分别对源图像作小波分解。融合图像的低频系数由一组区域性能评价参量按隶属度控制规则自适应调节得到;高频系数则直接采用可见光图像的高频系数,对新的低频和高频系数实施小波反变换即可得到融合图像。实验结果表明该方法能够取得很好的融合效果。
In order to consider image fusion evaluation information when choose a fusion method, a self-adaptive fusion method for SAR and optical images based on regional performance was proposed. Firstly, a smooth filter template was used to suppress the speckle of SAR image. Then the wavelet transform was used to perform a mttltiscale decomposition of each image, To low-frequency images, by extracting a set of regional performance parameters respectively, a self-adaptive fusion was applied with the regulation of membership degree; the fusion of high-frequency images adopted high-frequency of the optical image directly. The result image was generated by reverse wavelet transform, Experiment results show that this method can obtain satisfactory effect.
出处
《计算机应用》
CSCD
北大核心
2005年第11期2595-2597,共3页
journal of Computer Applications
基金
国家863计划项目(2002AA783055)
关键词
图像融合
小波变换
区域性能
自适应加权
image fusion
wavelet transform
region performance
self-adaptive weighting
作者简介
（susie@mail．ustc．edu．cn）刘媛（1982-），女，安徽芜湖人，硕士研究生，主要研究方向：图像处理、模式识别；
尹东（1965-），男，江西莲花人，副教授，主要研究方向：模式识别、图像分析与理解、数据库应用．
