融合光谱和空间信息的黄河主溜特征提取
Main-Stream Feature Extraction of the Yellow River Fusing Spectral and Spatial Feature
-
摘要: 本文基于多光谱遥感图像, 提出了一种融合光谱特征和空间信息的主溜特征提取方法.将遥感影像的两大表现形式——光谱和空间信息结合, 利用像元及其邻域的关系来描述其空间结构, 用方向线来描述黄河主溜, 从而确定其位置和走向.实验结果表明, 该方法可有效地提取出黄河主溜特征.Abstract: A new mainstream feature extraction approach is proposed, which efficiently fuses the two main manifestations in remote sensing images: spectral features and spatial information.The approach depicts the spatial structures with the relationships between pixels and their neighborhoods, and describes the main-stream with the direction line.Finally, the position and trend of the main-stream line are achieved.The experiments results show that the presented approach is competent for main-stream feature extraction and the feature could be extracted effectively.