城市绿地信息提取中高分辨率卫星影像融合方法研究

利用GS变换、主成分分析、Ehlers变换、Wavelet分析、HIS变换5种方法对城区WorldView-2和PL-1A影像进行融合，并从影像融合质量和绿地信息提取精度两方面对融合方法的有效性进行了评价。结果表明：①5种融合方法中，GS变换融合的效果最好；主成分分析和Ehlers变换融合WorldView-2质量较好，但融合PL-1A影像质量较差；Wavelet变换、HIS变换融合两种影像质量都较差；②用于绿地信息提取时，GS、PCA融合影像获取的精度最高，其次为Ehlers、Wavelet融合影像，均明显高于多光谱影像的提取精度；Ehlers、Wavelet变换精度最低，绿地信息提取精度低于多光谱影像的提取精度。可以得出，影像融合可以明显地提高绿地信息提取精度，5种影像融合方法中，GS变换普适性较好，影像融合质量最好，提高分类精度效果最明显。

Study of Fusion Algorithms with High Resolution Remote Sensing Image for Urban Green Space Information Extraction

High remote sensing image has been applied in extracting urban green space information widely, however effectiveness of fusion algorithms need further research.GS, PCA, Ehlers, Wavelet, HIS are applied to merger of urban WorldView-2 and PL-1A image.Then the effectiveness of fusion algorithms are evaluated according to quality of fusion and accuracy of urban green space information extraction.The results show that GS fusion algorithm achieves the best results.Quality of WorldView-2 fusion images using PCA and Ehlers fusion algorithms is better than that of PL-1A fusion images applied same algorithms.Wavelet and HIS are worst in five algorithms.Accuracy of urban green space extraction based GS, PCA, Ehlers, and Wavelet fusion images are better than accuracy of multispectral images.In those four fusion algorithms, GS and PCA are best while Ehlers and Wavelet are slightly inferior.From what has been discussed above shows that fusion algorithms can improve the accuracy of green space information extraction obviously.The GS algorithm has best fusion result and good universality in five fusion algorithms.