Please wait a minute...
img

官方微信

遥感技术与应用  2012, Vol. 27 Issue (4): 600-608    DOI: 10.11873/j.issn.1004-0323.2012.4.600
遥感应用     
影响ETM影像土地利用/覆盖分类精度因素的研究
赵 慧1,2,汪云甲1,2
(1.中国矿业大学环境与测绘学院,江苏 徐州 221116;2.中国矿业大学国土环境与灾害监测国家测绘局重点实验室,江苏 徐州 221116)
Research on the Factors Affecting the Classification Accuracy of ETM Remote Sensing Image Land Cover/Use
Zhao hui1,2,Wang Yunjia1,2
(1.School of Environment Science and Spatial Information,China University of Mining and Technology,Xuzhou 221116,China;
2.Key Laboratory for Land Environment and Disaster Monitoring of SBSM,China University of Mining and Technology,Xuzhou 221116,China)
 全文: PDF(5205 KB)  
摘要:

训练样本量、辅助数据和分类法是影响土地利用/覆盖分类精度的3个主要因素,通过找到这3个因素的最佳组合方式以提高分类精度,分别在25%、50%、75%、100%样本量下,加入NDVI、DEM和纹理均值特征作为辅助数据,比较了分类回归树、支持向量机、最大似然法3种分类法的效果,探讨了训练样本、辅助数据以及分类技术对土地利用/覆盖分类精度的影响。结果表明:支持向量机总体分类精度较高,在相同样本量和没有有效辅助数据的情况下,SVM可以获得最佳的分类结果,总体分类精度在85%以上;在进行分类时,加入NDVI和纹理均值特征使分类回归树分类精度提高了2.82%,说明该方法对有效辅助数据的加入较为敏感;在获取的训练样本集有限而可获取有效的辅助数据时,应优先考虑利用分类回归树进行土地利用/覆盖分类。

关键词: 决策树支持向量机训练样本量土地利用/覆盖分类    
Abstract:

Training sample size,auxiliary data and classification algorithm are the three main factors influencing on the land use and cover classification accuracy,through finding out the best combination of the three factors to improve the classification accuracy.Multispectral imagery and DEM from ETM satellite are applied to perform experiments under four sample sizes (100%,75%,50%,25%),along with calculated NDVI layer and mean texture measure.By contrasting the classification accuracy produced using three different classification algorithms:Classification and Regression Tree(CART),Support Vector Machine(SVM) and Maximum Likelihood Classification(MLC),to discuss the influence of training sample,auxiliary data and different classification algorithms to the classification accuracy.The results indicate that the SVM can obtain the best classification result when there is no effective auxiliary data,and the total accuracy can reach to 85%,when CART is used to classification ,as the additional NDVI and texture feature,the accuracy increased 2.82%,it is sensitive to auxiliary data ,which can obviously improve the classification accuracy.While in the condition of limited training sample obtain the effective auxiliary data,the CART can achieve the highest classification accuracy.

Key words: CART    SVM    Training sample    Land use/cover classification
收稿日期: 2011-12-18 出版日期: 2012-08-24
:  TP 79  
基金资助:

资源三号卫星立体测图技术与应用示范项目(2011BAB01B06\|06);江苏高校优势学科建设工程资助项目(PAPD,SA1102)。

作者简介: 赵 慧(1987-),女,河北泊头人,硕士研究生,主要从事遥感土地利用/覆盖分类研究。Email:charis_zh@yahoo.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  

引用本文:

赵 慧,汪云甲. 影响ETM影像土地利用/覆盖分类精度因素的研究[J]. 遥感技术与应用, 2012, 27(4): 600-608.

Zhao hui,Wang Yunjia. Research on the Factors Affecting the Classification Accuracy of ETM Remote Sensing Image Land Cover/Use. Remote Sensing Technology and Application, 2012, 27(4): 600-608.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2012.4.600        http://www.rsta.ac.cn/CN/Y2012/V27/I4/600

[1]Muchoney D,Borak J,Borak H C,et al.Application of the MODIS Global Supervised Classification to Vegetation and Land Cover Mapping of Central America[J].International Journal of Remote Sensing,2000,(21):1115-1138.
[2]Hui Wenhua.TM Image Classification based on Support Vector Machine[J].Journal of Earth Sciences and Environment,2006,28(2):93-95.[惠文华.基于支持向量机的遥感图像分类方法[J].地球科学与环境学报,2006,28(2):93-95.]
[3]Shao Xiaomin,Liu Yong.Deriving Vegetation Information in Ulan Buh Desert based on Texture[J].Remote Sensing of Technology and Application,2010,25(5):687-694.[邵晓敏,刘勇.基于纹理的乌兰布和沙漠地区植被信息提取[J].遥感技术与应用,2010,25(5):687-694.]
[4]Peddle D R,Franklin S E.Image Texture Processing and Data Intergration for Surface Pattern Discrimination[J].Photogrammetric Engineering and Remote Sensing,1991,57(4):413-420.
[5]Zhang Jinshui,He Chunyang,Pan Yaozhong,et al.The High Spatial Resolution RS Image Classification based on SVM Method with the Muti-Source Data[J].Journal of Remote Sensing,2006,10(1):49-57.[张锦水,何春阳,潘耀忠,等.基于SVM的多源信息复合的高空间分辨率遥感数据分类研究[J].遥感学报,2006,10(1):49-57.]
[6]Yan Jing,Wang Wen,Li Xiangge.Extracting the Rice Planting Area Using an Artificial Neural Network[J].Journal of Remote Sensing,2001,5(3):227-231.[阎静,王汶,李湘阁.利用神经网络方法提取水稻种植面积—以湖北省双季早稻为例[J].遥感学报,2001,5(3):227-231.]
[7]Zhu Xiufang,Pan Yaozhong,Zhang Jinshui,et al.The Effects of Training Sample on the Wheat Planting Area Measure Accuracy in TM Scale(iv):The Accuracy Response of Different Classifiers to Training Samples[J].Journal of Remote Sensing,2007,11(6):826-837.[朱秀芳,潘耀忠,张锦水,等.训练样本对TM尺度小麦种植面积测量精度影响研究(iv)-训练样本与分类方法间分类精度响应关系研究[J].遥感学报,2007,11(6):826-837.]
[8]Zhao Ping,Fu Yunfei,Zheng Liugen,et al.CART-based Land Use/cover Classification of Remote Sensing Images[J].Journal of Remote Sensing,2005,9(6):708-716.[赵萍,傅云飞,郑刘根,等.基于分类回归树分析的遥感影像土地利用/覆盖分类研究[J].遥感学报,2005,9(6):708-716.]
[9]Li Shuang,Zhang Erxun.The Decision Tree Classification and Its Application in Land Cover[J].Area Research and Development,2003,22(1):17-21.[李爽,张二勋.基于决策树的遥感影像分类方法研究[J].地域研究与开发,2003,22(1):17-21.]
[10]Liu Jianghua,Cheng Junshi,Chen Jiapin.Support Vector Machine Training Algorithm:A Review[J].Information and Control,2002,31(1):45-50.[刘江华,程君实,陈佳品.支持向量机训练算法综述[J].信息与控制,2002,31(1):45-50.]

[1] 王常颖,田德政,韩园峰,隋毅,初佳兰. 基于属性差决策树的全极化SAR影像海冰分类[J]. 遥感技术与应用, 2018, 33(5): 975-982.
[2] 熊伟,徐永力,姚力波,崔亚奇. 基于SVM的高分辨率SAR图像舰船目标检测算法[J]. 遥感技术与应用, 2018, 33(1): 119-127.
[3] 王凯,赵军,朱国锋. 基于GF-1遥感数据决策树与混合像元分解模型的冬小麦种植面积早期估算[J]. 遥感技术与应用, 2018, 33(1): 158-167.
[4] 周晓宇,陈富龙. 四川大熊猫栖息地PALSAR时序数据森林覆盖动态监测研究[J]. 遥感技术与应用, 2017, 32(6): 1100-1106.
[5] 陈洋波,张涛,窦鹏,董礼明,陈华. 基于SVM的东莞市土地利用/覆被自动分类误差来源与后处理[J]. 遥感技术与应用, 2017, 32(5): 893-903.
[6] 姬忠林,张月平,李乔玄,刘绍贵,李淑娟,任红艳. 基于GF-1影像的冬小麦和油菜种植信息提取[J]. 遥感技术与应用, 2017, 32(4): 760-765.
[7] 吕利利,颉耀文,黄晓君,张秀霞,李汝嫣. 基于CART决策树分类的沙漠化信息提取方法研究[J]. 遥感技术与应用, 2017, 32(3): 499-506.
[8] 史飞飞,高小红,杨灵玉,何林华,贾伟. 基于HJ-1A高光谱遥感数据的湟水流域典型农作物分类研究[J]. 遥感技术与应用, 2017, 32(2): 206-217.
[9] 胡根生,吴问天,黄文江,梁栋,黄林生. 粒子群优化的最小二乘支持向量机在小麦白粉病监测中的应用[J]. 遥感技术与应用, 2017, 32(2): 299-304.
[10] 郝泷,陈永富,刘华,朱雪林,达哇扎西,李伟娜. 基于纹理信息CART决策树的林芝县森林植被面向对象分类[J]. 遥感技术与应用, 2017, 32(2): 386-394.
[11] 鲍蕊,夏俊士,薛朝辉,杜培军,车美琴. 基于形态学属性剖面的高光谱影像集成分类[J]. 遥感技术与应用, 2016, 31(4): 731-738.
[12] 雷光斌,李爱农,谭剑波,张正健,边金虎,靳华安,赵伟,曹小敏. 基于多源多时相遥感影像的山地森林分类决策树模型研究[J]. 遥感技术与应用, 2016, 31(1): 31-41.
[13] 董保根,马洪超,车森,解龙根,何乔. LiDAR点云支持下地物精细分类的实现方法[J]. 遥感技术与应用, 2016, 31(1): 165-169.
[14] 林志垒,晏路明. 高光谱影像的BDT-SVM地物分类算法与应用[J]. 遥感技术与应用, 2016, 31(1): 177-185.
[15] 韩忻忆,颉耀文. 基于决策树的干旱区湿地信息自动提取—以疏勒河流域为例[J]. 遥感技术与应用, 2015, 30(6): 1146-1152.