基于Contourlet区域对比度的遥感图像融合算法

为综合利用多传感器遥感图像之间的信息，提出了一种基于contourlet变换的遥感图像融合算法。Contourlet是近年发展起来的一种多尺度几何分析工具，比小波变换更适合对图像等二维信号进行处理。算法的思想是将源图像分别进行contourlet变换，在contourlet域完成图像信息的融合，最后通过逆变换得到融合结果。根据分解系数特性，首次提出了contourlet域的区域对比度作为高频子带的融合规则，系数选择更符合人类视觉系统的要求，使得融合图像更有意义。低频子带采用基于区域能量的加权平均作为系数融合规则，在充分保留图像主要信息的前提下进一步增强了算法的稳定性。实验结果表明该算法在性能上明显优于PCA及小波图像融合等传统算法，且能够有效增强源图像的对比度和细节信息。

Data Fusion of Multisensor Remote Sensing Images using Region Based Contourlet Contrast

Remote sensing image fusion aims at processing and synthesizing information provided by various sensors.In this paper,the application of the contourlet transform in image fusion was discussed.Contourlet is a novel multiscale geometric analysis tool;it provides many advantages in comparison with conventional multiresolution analysis.The contourlet contrast measurement was first developed,which is proved to be more suitable for image processing tasks.Then a novel image fusion algorithm using region based contourlet contrast was proposed.The fusion rules also adopted the local energy and the "weighted averaging" method,which can preserve more details in source images and further improve the subjective quality of fused image.Experimental results show that comparing with traditional image fusion algorithms,the proposed approach can provide more satisfactory fusion outcome.