采用基于数据并行的遥感影像分割实现过程, 提出了一种新的数据缝合算法解决分割结果合并问题。分割效果对比和运算效率分析等实验结果表明, 此算法保持了分割结果合并后的边界正确性, 使并行化分割在提高运算效率的同时保证了分割结果的可信度。

In the process and analysis of high-resolution remote sensing image, segmentation is the key step of extracting information from image data to image object. For the image segmentation tasks of large amount of data, data paralleled computing model is generally used. In this process, the effect of merging segmentation results when data gathering is related to the precision and accuracy of the subsequent object-oriented analysis. In this paper, data paralleled segmentation of remote sensing image is adopted, and a new algorithm named data sewing is proposed to solve the problem of merging segmentation results. Experi-ments, such as comparison of final segmentation results and assessment of computing efficiency, show that the algorithm im-proves the efficiency of image segmentation process. Meanwhile it guarantees the correctness of the boundary thus to ensure the credibility of the final segmentation result as well.