一种改进的遥感影像云检测方法  

An Improved Cloud Detection Method in Remote Sensing Imagery

作　　者：张昊[1] 焦瑞莉[1] 乔聪聪 霍娟[2] 宗雪梅[2] ZHANG Hao;JIAO Ruili;QIAO Congcong;HUO Juan;ZONG Xuemei(School of Information Communication Engineering,Beijing Information Science and Technology University,Beijing 100101,China;Institute of Atmospheric Physics,Chinese Academy of Sciences,Beijing 100029,China)

机构地区：[1]北京信息科技大学信息与通信工程学院,北京100101 [2]中国科学院大气物理研究所,北京100029

出　　处：《遥感信息》2023年第3期130-137,共8页Remote Sensing Information

基　　金：国家自然科学基金面上项目(42175150)。

摘　　要：针对多特征联合方法不能有效滤除与云具有相似几何特征的非云区域,以及其滤除冰雪区域存在局限性的问题,提出两点改进,并采用波段范围更加丰富的Landsat 8遥感影像验证此方法在其他遥感数据上的性能。改进方法利用灰度共生矩阵提取纹理特征,并结合原始方法应用的几何特征进一步滤除具有云相似几何特征的非云区域;引入了一种光谱特征对冰雪区域进行滤除,解决了原始方法的局限性。定量分析与实验结果表明:与原始方法相比,改进方法的云检测结果中对非云区域滤除效果更佳,云检测精度更高。To address the limitations of the multi-feature combined method in effectively filtering out non-cloud areas with similar geometric features to clouds and filtering out snow and ice areas,two improvements are proposed,and the performance of the method on other remote sensing data is verified by using Landsat 8 remote sensing images with a richer waveband range.The improved method uses the gray level co-occurrence matrix to extract texture features,and combines the geometric features applied by the original method to further filter out non-cloud areas with similar geometric features to clouds.A spectral feature is introduced to filter out ice and snow areas,addressing the limitations of the original method.Quantitative analysis and experimental results show that compared with the original method,the cloud detection results of the improved method have better filtering effect on non-cloud areas and higher cloud detection accuracy.

关 键 词：多特征联合 Landsat 8 云检测 灰度共生矩阵 纹理特征 几何特征 光谱特征 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]