作者: Ye Zhang , Yijia Liu , Junping Zhang
关键词:
摘要: In view of the problem non-target interference and misjudgment in road extraction from remote sensing images, a new method based on curvelet fusion is presented. Firstly spectral similarity matrix calculated multispectral images. Then panchromatic image are fused by using transform. The result processed with cross-correlation line detection isolated pixels suppression. Finally features extracted Hough order to verify effectiveness algorithm, experiments performed many different scenes. results show that has good robustness accuracy. Roads can be accurately both simple complex