针对利用常规方法对多光谱图像与全色图像融合过程中出现的颜色失真问题,提出了一种基于区域颜色校正的遥感图像融合方法。该方法首先利用模糊C-均值(FCM)聚类算法在多特征形成的特征空间上对图像进行区域分割,然后利用常规方法融合多光谱图像与全色图像,并采用小波变换对融合图像进行区域上的颜色校正,从而得到最终的融合图像。通过与IHS、BT和PCA几种常规融合方法的比较与分析验证了本文提出的方法具有良好的融合特性。
In order to reduce the color distortion existing in the fusion image generated from the multispectral image and the panchromatic image,a method about remote sensing image fusion based on regional color correction is presented. First, the FCM(Fuzzy C-Means) clustering algorithm is used to segment the image in the feature space formed by the multiple features of training samples. Second,the general method is used to fuse multi-spectral image and panchromatic image. The fused image is obtained by using the wavelet transform to modify the regional color of the fusion image. The effectiveness of the proposed image fusion method is verified by contrast to IHS,BT and PCA fusion methods.