ASTER多光谱影像与资源二号全色影像融合研究

Research on Fusion of ASTER Multi-spectral Image and Resources-2 Panchromatic Image

随着遥感获取数据手段的日益增多,遥感影像的融合技术备受关注。针对ASTER多光谱影像光谱信息丰富、分辨率小,资源二号全色影像分辨率高、纹理信息丰富的特点,以安徽省马鞍山市当涂县的ASTER多光谱影像与资源二号全色影像为例,分别采用基于主成分分析、小波变换以及主成分分析与小波变换相结合的3种融合方法进行融合实验,并对融合后的影像进行对比,探讨ASTER多光谱影像与资源二号全色影像融合的方法和效果。结果表明:采用主成分分析与小波变换相结合的方法对两幅影像融合的效果最好,极大地改善了两种单一方法的缺点,提高了原始影像的目视效果和光谱信息,从而为区域研究提供了更精确的数据资料。

Remote sensing image fusion technology is concerned as the ways to obtain the RS data are growing. Aiming at the characteristics of multi-spectral imaging with more spectral information, Low-resolution and Resource-2 panchromatic images with high-resolution, rich texture information, this article, taking the ASTER multi-spectral image and Resources-2 panchromatic image of Dangtu County in Ma＇anshan,Anhui Province as a case, discusses the fusion methods and effects of two kinds of images using three fusion methods based on principal component transformation, wavelet transform as well as principal component transformation combination with wavelet transform respectively, and comparing three fused images. The results show that the method based on principal component transformation combination with wavelet transform owns the best integration effect. It greatly overcomes the shortcoming of the fusion methods based on single transformation,and improves the visual effects and spectral information of the original image to provide more precise data for regional research.