基于影像特征融合的遥感图像重构模式研究

基于影像特征级数据融合的遥感图像重构是在突出目标地物的空间结构和纹理特征情况下的信息融合。在数字图像小波多分辨率分析理论基础上，采用小波变换方法对高分辨遥感图像的目标地物边缘进行信息增强，然后与多光谱遥感图像进行特征信息融合。在融合过程中，首先对多光谱图像中的R、G、B三个波段的图像进行小波分解，得到相应的低频图像，然后对特征增强后的高分辨率图像进行小波分解，再将分解后的高频图像分别与低频图像进行融合，最后经RGB合成为彩色图像。该方法既改善了图像的清晰度和分辨率，同时也保留了原图像的光谱信息。通过融合实验验证了上述结论。

Research on model of remote sensing image rebuilding based on image feature fusion

Remote Sensing(RS) image rebuilding based on image feature Level data fusion is a sort of approach based on spectral structure and texture feature of the objects. After the wavelet's multi-resolution analysis, a feature fusion approach was adopted to enhance remote sensing image edge and improve the definition and resolving power of the image. In the fusion process, instead of using the conventional HIS-Wavelet fusion, this approach used low frequency parts of the R, G, B bands of the multi-spectral image decomposed by the wavelet and high frequency parts of the high-resolution image also decomposed by the wavelet to make the image fusion separately, and finally the new fusion image was composed by the new R, G, B bands. In this way, this approach not only improved the definition and resolution of the multi-spectral image, but also retained the color feature which was important for multi-spectral fusion. In the end, the conclusion mentioned above was validated with an experiment.