Hyperspectral change detection based on independent component analysis
基于独立成分分析的高光谱变化检测

Change detection is the process of analyzing changes of surface features with multi-temporal remote sensing imagery of the same area. Hyperspectral remote sensing images contain abundant spectral information for accurate change detection, which, regrettably, is not fully taken into account by existing approaches. In this paper, a hyperspectral change detection method based on Independent Component Analysis (ICA) is investigated. The difference image is analyzed by skew-based ICA. The change of a single feature can be obtained and then the change is extracted from each abundance image. Experiment results demonstrate that the ICA-based hyperspectral change detection performs better than other traditional methods with a high detection rate and a low false detection rate.