您的位置:首页  > 论文页面

基于目标特征的遥感图像分类方法进展

发表时间:2009-11-15  浏览量:1368  下载量:675
全部作者: 王珂,周坚华
作者单位: 华东师范大学资源与环境科学学院;华东师范大学地理信息科学教育部重点实验室
摘 要: 根据遥感图像分类识别的一般步骤(图像选择与预处理、目标特征提取、分类方法选择及分类结果输出),综合分析了遥感图像的目标特征(光谱特征、纹理特征、空间特征和多特征)提取算法。描述了基于目标特征的各种分类识别方法并进行了归纳总结。最后,指出了上述方法的不足并对遥感图像识别分类的研究发展进行了展望。
关 键 词: 地理学;遥感图像分类识别;综述;特征提取;支持向量机;人工神经网络
Title: Progress of remote sensing image classification methods based on target characteristics
Author: WANG Ke, ZHOU Jianhua
Organization: School of Resources and Environment Science, East China Normal University;Key Laboratory of Geographic Information Science, Ministry of Education, East China Normal University
Abstract: According to the general steps of remote sensing image classification recognition, which include image selection and pre-processing, target feature extraction, classification method selection and classification result output, this paper comprehensively analyzes characteristic (such as spectral characteristic, texture characteristic, spatial characteristic and multi-characteristic) extraction algorithms of remote sensing image. It also describes and summarizes various classification recognition methods based on target characteristics, and expert system classification. At last, it points out the defects of the above-mentioned methods and forecasts the development of research about the remote sensing image classification recognition.
Key words: geography; classification recognition of remote sensing image; review; characteristic extraction; support vector machine; artificial neural networks
发表期数: 2009年11月第21期
引用格式: 王珂,周坚华. 基于目标特征的遥感图像分类方法进展[J]. 中国科技论文在线精品论文,2009,2(21):2236-2242.
 
0 评论数 0
暂无评论
友情链接