摘要
随着遥感成像技术的发展,高光谱图像的应用需求日益广泛。如何从多达数百个的波段中挑选出具有较好识别能力的波段组合成了亟待解决的问题。根据高光谱图像各波段间相关性高的特点,提出了基于条件互信息与自适应分支定界法相结合的波段分组方法,并在此基础上使用支持向量机和遗传算法相结合的搜索算法,选择最佳波段组合。实验结果表明:提出的算法具有相当出色的分类准确率和稳定性。
With the development and popularization of the remote-sensing imaging technology,there are more and more applications of hyperspectral image classification tasks,such as target detection and landcover investigation. It is a very challenging issue of urgent importance that how to select a minimal and effective subset from mass of bands. A novel band selection strategy is put forward based on conditional mutual information between adjacent bands and branch and bound algorithm for the high correlation between the bands. In addition,genetic algorithm and support vector machine are employed to search for the best band combination. Experimental results show that the proposed approach is very competitive and robust.
出处
《计算机科学与探索》
CSCD
2010年第5期464-472,共9页
Journal of Frontiers of Computer Science and Technology
基金
国家自然科学基金No.60673141
国家"十一五"科技支撑计划重大项目No.2006BAB04A13~~
关键词
高光谱遥感图像
波段选择
条件互信息
自适应分支定界法
支持向量机
遗传算法
hyperspectral remote sensing
band selection
conditional mutual information
adaptive branch and bound algorithm
support vector machine
genetic algorithm
