摘要
鉴于遥感影像上云的存在是影响影像质量判定的一个重要因素,对下垫面地物的识别与分类造成困难,降低了影像的利用率,分析了不同类型的云斑在资源三号卫星影像上的光谱亮度特征与直方图曲线分布特征,使用灰度共生矩阵提取云的纹理特征,依次构建云的光谱特征和纹理特征的判别准则,采用决策树分类法识别检测含云量,提出了一种针对资源三号卫星影像云量判读的算法。实验结果表明,该算法能够定性定量地判读出资源三号卫星影像上不同类型的云含量,正确率高达96%以上,具有效果好且速度快的特点。
The existence of clouds in remote sensing images is an important factor in determining image quality, causing difficulties on identification and classification of underlying surface features, and reducing image utilization. This algorithm analyzes the spectral brightness and histogram curve distribution characteristics of different types of clouds in ZY-3 images,and extracts the texture feature of cloud by GLCM,followed by building rules via cloud spectral and texture features,and using the decision tree classifier to detect clouds in ZY-3 images. The experimental results show that this algorithm can qualitatively and quantitatively interpret different types of clouds in ZY-3 images, and the correct rate is more than 96%, which has the characteristics of good effect and fast speed.
作者
刘志恒
韩玲
周平
王霞
吴婷婷
LIU Zhiheng HAN Ling ZHOU Ping WANG Xia WU Tingting(School of Geology Engineering and Geomatics, Chang＇an University, Xi＇an 710054, China Satellite Surveying and Mapping Application Center, National Administration of Surveying, Mapping and Geoin formation , Beijing 100048, China)
出处
《遥感信息》
CSCD
北大核心
2017年第4期41-46,共6页
Remote Sensing Information
基金
中央高校基本科研业务费专项资金资助项目(310826161009
310827162025)
国土资源部退化及未利用土地整治工程重点实验室开放基金(SXDJ2017-7)
