With the continuous improvement of the spatial resolution of remote sensing image, the interference of weak edges and false edges caused by the shadow to accurate extraction of the object outlines has been moreprotruding. Therefore, this paper introduces a shadow compensation strategy into image segmentation to propose a high resolution remote sensing image segmentation algorithm SWJSEG for urban scenes based on the previously proposed WJSEG algorithm. Firstly, the shadow is regarded as the point out of bounds to get the mean value and covariance matrix through the iterative strategy to realize the shadow detection and compensation based on Chisquare transformation. On this basis, the multiscale segmentation algorithm WJSEG is adopted to obtain the final segmentation result. The experimental results show that this method proposed in this papercan effectively deal with the interference caused by the shadow and improve the accuracy and reliability of image segmentation.