基于遥感数据特征和像元替换技术的边缘提取方法

随着城市化的发展，城市遥感领域在中国发展迅速。高分辨遥感图像的边缘提取是当前城市遥感的重要研究领域。实践中，传统的边缘检测算子主要是通过图像空城特征微分，建立不同结构的模板完成。高分辨率遥感数字图像包括了空间域和光谱域两种信息，因此，借助于图像的光谱和空间域两种特征信息提高提取城市边缘信息精度已经成为当前算法开发的基本思路。该研究通过设计光谱分解和边缘检测的算子模板，综合利用了图像的光谱、空间特征信息。研究结果表明，这种方法有效地提高了城市遥感数据边缘信息的提取精度，同时还具有方法简便、计算速度怏的特点。

Edge Extraction Method Based on Features of Remote Sensing Data and Pixel Swapping

With urban and township development in China, city remote sensing as base data has developed rapidly. The edge extraction of the remote sensing image of high resolution is the important research area of city remote sensing information extraction. Practically, it is competed in traditional edge detection kernel such as Sobel kernel through different template of differential coefficient in spatial territory. Spatial information as well as spectral information is included in the remote sensing image of high resolution .So, it will be an essential thoughtway of current method in city edge information extraction integrating spatial and spectral features of image. The processing is designed for edge information extraction with spectral analyzing, integrating spectral and spatial features of image. The results proves its effectiveness and shows that the precision of city remote sensing image edge information extraction is improved with the method , simultaneously, its an easy way to understand and have a high rate.