Change detection of buildings in high-resolution satellite images based on quantum multi-scale fusion
In order to improve the accuracy of the traditional high-resolution satellite image change detection method based on pixels, this paper proposes a building change detection algorithm based on quantum multi-scale fusion for high-resolution satellite images. Firstly, multi-scale segmentation of dual temporal high-resolution satellite images is carried out to form a multi-scale image dataset. Secondly, the multi-scale image dataset is transformed by iterative slow feature transformation to obtain the change intensity map of different scales, and then the multi-scale change intensity map is fused by quantum theory to obtain the fused change intensity map. Finally, the threshold segmentation of the change intensity map is completed by the maximum variance between classes method, and the binary change detection results are obtained. Two groups of real high-resolution satellite images with different time phases are used to verify the algorithm in this paper. The experimental results show that compared with the single-scale object-oriented change detection method and the multi-scale fusion method of entropy weight method, the algorithm in this paper can achieve higher accuracy in building change detection.
万方数据
维普
