Change Detection in High Spatial Resolution Images Based on Support Vector Machine

With the tendencies of "3-High" for RS images (high spatial resolution, high spectral resolution and high temporal resolution), more attentions are paid to the information processing technique for high-resolution image data. However, the performances of current high spatial resolution RS change detection methods and systems are not satisfying in both effect and efficiency. A new unified approach is presented that integrates SVM based classifier to change detection (SVMCCD). Combined with the change detection task, a bootstrapping strategy is proposed to solve sample selection problem. Considering the relative simplicity of non-change patterns, one- class SVM based change detection method is also provided.