论文部分内容阅读
一、引言遥感图像的边缘检测,在遥感技术应用研究中有着重要意义。本文是利用陆地卫星图像,根据土地资源分类调查的需要,提取图像的边缘信息。通过理论分析和实验证明,采用本文提出的处理方法可得到较好的边缘检测结果。图像边缘通常定义为局部区域内图像特性的差异,它表现为图像的不连续性(如灰度级、纹理的突变等)。在图像分割技术中,常用微分法进行边缘检测。通常选用一个阈值对梯度的幅值作二值化处理,以得到实际边缘。此时的困难是难以找到最佳阈值。这种
I. INTRODUCTION Edge detection of remote sensing images is of great significance in the application of remote sensing technology. In this paper, we use the land satellite imagery to extract the edge information of the image according to the needs of land resources classification survey. Through theoretical analysis and experiment, we can get a better edge detection result by using the method proposed in this paper. Edge of the image is usually defined as the difference of image characteristics in the local area. It shows the image discontinuity (such as gray level, texture mutation, etc.). In image segmentation technology, the differential method commonly used for edge detection. Usually choose a threshold for the magnitude of the gradient for binary processing, in order to get the actual edge. The difficulty here is that it is hard to find the optimal threshold. This