Due to the great difference of imaging methods between optical and SAR images, it is very difficult to extract a unified set of optical and SAR objects. Therefore, this paper proposes a geographic object-oriented integrated segmentation method of high resolution optics and SAR. This method is different from the traditional integrated segmentation strategy of processing heterogeneous images at the same time, but only segmenting optical images, so as to obtain a reliable set of geographical objects; On this basis, the marker points are extracted adaptively for each object, and projected into the SAR image according to the coarse registration results; Finally, the region growth based on marker points is carried out in SAR image, and finally the object set matching the segmented object of optical image is obtained. The experimental results of several groups of optical and SAR images show that the optical SAR matching object set extracted by the proposed method is closer to the actual geographical object, and the j-value can reach more than 7.8, which is significantly better than the comparison method in visual analysis and quantitative evaluation.