基于PCA、SFIM、MLT算法的ETM+影像融合

Fusion of ETM+ Images Based on PCA,SFIM and MLT Algorithms

采用PCA、SFIM和MLT3种融合算法对ETM+影像进行融合,并从融合影像的光谱质量、空间结构信息和分类精度等方面对融合方法进行评价.结果表明,SFIM融合法的光谱质量最高,PCA融合法具有最高的高频空间结构信息融入度,MLT融合法具有最高的分类精度,3种融合影像的分类精度都较原始影像的分类精度有所提高.

Three image fusion algorithms are employed to fuse ETM ＋ panchromatic and multispectral images to achieve both high spatial and spectral resolution in a single image. These are the principal component analysis （PCA）, smoothing filter-based intensity modulation （ SFIM）, and multiplication transform （MLT）. The effectiveness of the three algorithms is evaluated on spectral fidelity, high spatial frequency information gain, and classification accuracy. The study reveals that SFIM transform is a good method in retaining spectral information of original image. PCA fused image has high frequency information gain, and MLT transform has high elassification accuracy. All of the fused images have better classification accuracy than the original one.