基于LSSVM和多数投票原则的遥感影像道路提取  

Road extraction of remote sensing image based on LSSVM and majority voting algorithm

作　　者：刘红彬[1,2] 楚恒 李洪川[1,2] 

机构地区：[1]重庆邮电大学通信与信息工程学院,重庆400065 [2]重庆高校市级光通信与网络重点实验室,重庆400065 [3]西南大学地理科学学院,重庆400715

出　　处：《电视技术》2017年第7期152-156,共5页Video Engineering

基　　金：重庆市2013西南大学博士后科研项目(Rc201336)

摘　　要：由于高分辨率遥感影像中道路信息的复杂性及其与其他地物光谱信息的相似性,导致基于像素道路提取方法中的道路特征利用不够充分。针对这个问题,对原始影像进行HIS变换提取光谱饱和度(Saturation,S)分量,运用基于区域邻接图(Region Adjacency Graph,RAG)的优化分割算法,实现道路区域的分割,然后利用基于最小二乘支持向量机(LSSVM)和多数投票原则对道路信息进行提取。实验表明,上述方法能有效地提取遥感影像中的道路信息。Because of the complexity of road information in the high resolution remote sensing image and the similarity of the spectral information of other features,the road feature extraction method based on the pixel road extraction method is inadequate. To solve this problem,the HIS transform is performed on the original image to extract spectral saturation（ Saturation,S） components.To realize road segmentation,we use the optimal segmentation algorithm which based on region adjacency graph（ Region Adjacency,Graph,RAG） to realize road segmentation,and then extract road information by the method which based on least squares support vector machine（ LSSVM） and majority voting principle. Experimental results show that the method can effectively extract the road information in remote sensing image.

关 键 词：HIS变换 区域分割 最小二乘支持向量机 多数投票 

分 类 号：TN39[电子电信—物理电子学]