中文摘要：

面向对象的遥感信息提取，首要任务是对遥感影像进行分割，分割的目的在于把原始图像分割成一些在空间上相邻、光谱相似的同质区域。传统的分水岭变换对噪声敏感，易产生过分割现象，不能直接适用于遥感影像的信息提取。本文利用非线性同组滤波方法（PGF）消除原始影像噪声，根据高分辨率遥感影像地物的异质性特征，对分水岭初始分割结果进行异质性准则合并。实验结果表明，该方法对高分辨率遥感影像的分割效果良好，能够适用于面向对象的信息提取。

英文摘要：

This paper used nonlinear filtering method of PGF to eliminale image noise, and then merged the initial results of watershed transformation based on the heterogeneity of lhe bigh-resolution remote sensed image features. Results of experiments demonstrated that this method would be good for the high-resolution remote sensed image segmentation and could be applied in object-oriented information extraction.