摘要
从传统的梯度算子的边缘检测算法对高分辨率遥感图像无法有效检测出发,探讨一种基于傅里叶变换和模糊推理的高分辨率遥感图像边缘特征检测的方法。该方法首先提取前n次谐波能量,根据直流中心频谱图像良好的噪声抑制特性,分析高频分量对图像边缘特征的贡献。然后采用巴特沃斯低通滤波器获取低频分量,采用模糊推理的方法提取低频分量中的边缘特征,将这两部分的边缘特征叠加,得到最终的边缘检测图像。仿真实验表明,所提出的算法检测出的图像边缘特征效果显著,抗噪能力强。
Traditional edge detection method with gradient operator can not effectively detects high-resolution remote sensing images, proceeding from this, we discuss a novel edge feature detection algorithm of high-resolution remote sensing images which is based on Fourier transform and fuzzy reasoning. First, the method extracts the first n-th harmonic energy, and analyses the contribution of higher frequency components on image edge feature according to the good noise suppression capacity of DC （direct current） spectrum image. Then it uses Butterworth low-pass filter to obtain low frequency component, and extracts edge feature in it using fuzzy reasoning method. Finally, these two parts of edge features are superimposed to get final edge detection image. Simulation experimental results show that the image edge features detected by the proposed algorithm has significant effect and high anti-noise ability.
出处
《计算机应用与软件》
CSCD
北大核心
2014年第10期227-230,共4页
Computer Applications and Software
关键词
边缘特征
谐波能量
高分辨率遥感图像
特征检测
Edge feature
Harmonic energy
High-resolution remotely sensed image
Feature detection
