Please wait a minute...
img

官方微信

遥感技术与应用  2013, Vol. 28 Issue (2): 263-268    DOI: 10.11873/j.issn.1004-0323.2013.2.263
图像与数据处理     
基于边缘特征和AdaBoost分类的遥感图像云检测
王奎1,张荣1,尹东1,张海堂2
(1.中国科学技术大学电子工程与信息科学系,安徽 合肥 230027;
2.西南电子电信技术研究所,四川 成都 610041)
Cloud Detection for Remote Sensing Image based on Edge Features and AdaBoost Classifier
Wang Kui1,Zhang Rong1,Yin Dong1,Zhang Haitang2
(1.Department of Electronic Engineering and Information Science,University of Science and
Technology of China,Hefei 230027,China;2.The South-west Institute of Electronics &
Telecommunication Technology,Chengdu 610041,China)
 全文: PDF(1780 KB)  
摘要:

传统遥感图像云检测方法在处理山地、雪地、暗云等场景时极易发生错判,准确度较低。通过对遥感图像中云与地物的不同特点进行分析,提出一种新的遥感图像边缘特征描述方法,结合图像的边缘特征和灰度特征使用AdaBoost分类器进行云图分类,并利用图像的空间相关性对分类结果进行修正。经10万余幅图像测试结果表明:该算法与传统算法相比准确度极大提高,正确率达到96%以上,且运算速度快,满足实时性要求。

关键词: 云检测边缘特征AdaBoost;邻域修正    
Abstract:

Cloud cover is an important factor that degrades the quality of remote sensing images.Generally,traditional cloud detection algorithms can not work effectively in scenes such as mountains,snow and dark clouds,thus resulting in lower detection precision.In this paper,we analyzed the difference of characteristics between cloud area and earth object,then defined a new edge feature descriptor,and lastly proposed a new cloud detection algorithm based on image block classification.The edge features and gray features are extracted and classified by the AdaBoost classifier,and the result is corrected by using spatial neighbouring correlation.More than 100 thousand image blocks experiment shows that this algorithm has much better performance than traditional algorithms.To be specific,the accuracy is more than 96% and the operation is so fast that real-time requirement can be ensured.

Key words: Cloud detection    Edge feature    AdaBoost    Neighbouring correction
收稿日期: 2012-02-22 出版日期: 2013-06-24
:  TP 79  
作者简介: 王奎(1986-),男,河南漯河人,硕士研究生,主要从事遥感图像处理方面的研究。Email:wangkui@mail.ustc.edu.cn。
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
王奎
张荣
尹东
张海堂

引用本文:

王奎,张荣,尹东,张海堂. 基于边缘特征和AdaBoost分类的遥感图像云检测[J]. 遥感技术与应用, 2013, 28(2): 263-268.

Wang Kui,Zhang Rong,Yin Dong,Zhang Haitang. Cloud Detection for Remote Sensing Image based on Edge Features and AdaBoost Classifier. Remote Sensing Technology and Application, 2013, 28(2): 263-268.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2013.2.263        http://www.rsta.ac.cn/CN/Y2013/V28/I2/263

[1]Zhang Bing.Intelligent Remote Sensing Satellite System[J].Journal of Remote Sensing,2011,15(3):415-431.[张兵.智能遥感卫星系统[J].遥感学报,2011,15(3):415-431.]
[2]Yu Fan,Chen Weimin.Research on the Cloud Classification for the Bi-spectrum Cloud Picture[J].Journal of Nanjing Institute of Meteorology,1994,17(1):117-124.[郁凡,陈渭民.双光谱云图的云分类探讨[J].南京气象学院学报,1994,17(1):117-124.]
[3]Ackerman S A,Strabala K I,Menzel W P,et al.Discriminating Clear Sky from Clouds with MODIS[J].Journal of Geophysical Research,1998,103(D24):32,141-32,157.
[4]Liu Xiantong,Liu Qi,Fu Yunfei,et al.Daytime Cloud Detection Scheme Relying on Five-channel Measurements from TRMM VIRS[J].Journal of Atmospheric and Environmental Optics,2010,5(2):128-140.[刘显通,刘奇,傅云飞,等.基于TRMM VIRS可见光和红外五通道的白天云检测方案[J].大气与环境光学学报,2010,5(2):128-140.]
[5]Chen Peng,Zhang Rong,Liu Zhengkai.Feature Detection for Cloud Classification in Remote Sensing Images[J].Journal of University of Science and Technology of China,2009,39(5):484-488.[陈鹏,张荣,刘政凯.遥感图像云图识别中的特征提取[J].中国科学技术大学学报,2009,39(5):484-488.]
[6]Yan Yusong,Long Teng.Real-time Cloud Detection in Optical Remote Sensing Image[J].Transactions of Beijing Institute of Technology,2010,30(7):817-821.[闫宇松,龙腾.遥感图像的实时云判技术[J].北京理工大学学报,2010,30(7):817-821.]
[7]Shan Na,Zheng Tianyao,Wang Zhensong.High-speed and High-accuracy Algorithm for Cloud Detection and Its Application[J].Journal of Remote Sensing,2009,13(6):1147-1162.[单娜,郑天垚,王贞松.快速高准确度云检测算法及其应用[J].遥感学报,2009,13(6):1147-1162.]
[8]Zhao Min,Zhang Rong,Yin Dong,et al.Cloud Classification Algorithm for Optical Remote Sensing Image[J].Remote Sensing Technology and Application,2012,27(1):106-110.[赵敏,张荣,尹东,等.一种新的可见光遥感图像云判别算法[J].遥感技术与应用,2012,27(1):106-110.]
[9]Gonzalez R C,Woods R E.Digital Image Processing (Second.Edition)[M].London:Prentice Hall,2002:572-584.
[10]Freund Y,Schapire R E.Desicion-theoretic Generalization of On-line Learning and an Application to Boosting[J].Journal of Computer and System Sciences,1997,55(1):119-139.
[1] 陈伟民,张凌,宋冬梅,王斌,丁亚雄,许明明,崔建勇. 基于AdaBoost改进随机森林的高光谱图像地物分类方法研究[J]. 遥感技术与应用, 2018, 33(4): 612-620.
[2] 杨春燕, 陈圣波, 包书新, 宋金红. 长春地区云检测及云相态反演研究[J]. 遥感技术与应用, 2010, 25(2): 245-250.
[3] 曹广真,漆成莉,马 刚,张凤英,吴雪宝. FY-3A气象卫星VIRR云检测产品与IRAS的匹配[J]. 遥感技术与应用, 2008, 23(1): 89-92.
[4] 谈建国,周红妹,陆 贤,杨崇俊. NOAA卫星云检测和云修复业务应用系统的研制和建立[J]. 遥感技术与应用, 2000, 15(4): 228-231.