摘要
将决策树算法引入到遥感影像分类中,以提高分类的精度。首先对影像进行预处理,然后利用C5.0算法在分析地物光谱特征、纹理特征、归一化植被指数的基础上,自动提取分类规则,构建决策树,实现地物的自动分类。为验证该算法的有效性,选取西藏某地区TM影像作为实验数据,与监督分类的精度进行对比,实验结果表明,决策树分类方法能取得较好的分类效果。
In order to enhance terrain classification accuracy,this paper introduces decision tree algorithm into Tibet Remote Sensing image classification.The experiments use C5.0 algorithm to analyse of spectral characteristic curve,NDVI and texture features,and then construct decision tree to achieve automatic classification after pre-progressing of the images.In order to verify the effectiveness of the algorithm,this paper selects an area of Tibet TM image as the experimental data.In comparison with the supervised classification,the experimental results show that decision tree classification method can achieve better classification results.
出处
《测绘》
2011年第1期3-6,共4页
Surveying and Mapping
关键词
遥感影像
分类
决策树算法
Remote sensing image
Classification
Decision tree classification
