Change Detection Method of High Resolution Remote Sensing Image Based on D-S Evidence Theory Feature Fusion

Using high-resolution satellite image to detect change has been a hotspot in the field of remote sensing for long time series. The detection method combining feature extraction and machine learning could extract information effectively, but manual sample selection is huge workload wide range images, it also difficult ensure accuracy pre-detection using single difference image. Therefore, this paper, new put forward based on multi-feature fusion D-S evidence theory. In approach, texture calculated by structural similarity, because similarity plays great role detection, which was verified experiments. images features traditional spectral are fused theory, have fully utilized. Setting rules select samples with high confidence pixels, SLIC super-pixel segmentation applied order improve further credibility sample. Finally, selected optimization sent classifier training obtain final result. experimental results show that play very important theory effectively fuse detection. proposed good performance