针对图割算法实时性差的问题,提出一种基于四叉树与图割的遥感图像分割方法。用四叉树分割方法将遥感图像分割成多个区域,根据区域的邻接性合并相似小区域,结合区域间的灰度和位置信息,利用Normalized Cut方法在区域间进行划分,完成图像的分割。实验结果证明该方法能有效消除四叉树方法的过分割现象。 Aiming at the problem of poor real-time graph cut algorithm, a remote sensing image segmentation method based on quadtree and graph cut is proposed. A quadtree segmentation method is used to segment the remote sensing image into multiple regions. The similar small regions are merged according to the adjacency of the regions. The normalized cut method is used to divide the regions according to the gray level and position information of the regions. Experimental results show that this method can effectively eliminate the quadtree over-segmentation method.