收稿日期: 2002-10-09
网络出版日期: 2003-05-10
基金资助
国家科技攻关计划项目(2001DFBA0005);中国科学院知识创新工程重大项目--中国陆地和近海生态系统碳收支研究(KZCX1SW01)资助
Landuse Classification in Arid and Semi-Arid Areas Using CBERS-1 Imagery
Received date: 2002-10-09
Online published: 2003-05-10
以中巴资源卫星CBERS 1图像数据为信息源,分别采用最大似然法、BP神经网络和Fuzzy ARTMAP神经网络 3种分类器,以位于干旱区的中国新疆石河子地区为例,进行了土地利用计算机自动分类。结果认为,3种方法中以Fuzzy ARTMAP神经网络法分类精度最高,分别比最大似然法和BP神经网络法提高了 10.69%和 6.84%。同时也证实了CBERS 1图像在土地利用调查中的实用性
关键词： CBERS-1图像; BP神经网络; Fuzzy-ARTMAP神经网络; 土地利用分类
刘爱霞 , 刘正军 , 王长耀 , 牛铮 . 基于CBERS-1图像的干旱半干旱区土地利用分类[J]. 中国科学院大学学报, 2003 , 20(3) : 334 -340 . DOI: 10.7523/j.issn.2095-6134.2003.3.012
Discussed and analyzed results of different classification algorithms for land use classification in arid and semiarid areas using CBERS-1 image, Which in case of our study is Shihezi Municipality, Xinjiang Province.Three types of classifiers are included in our experiment, including the Maximum Likelihood classifier, BP neural network classifier and Fuzzy-ARTMAP neural network classifier.The classification results showed that the classification accuracy of Fuzzy-ARTMAP was the best among three classifiers, increased by 10.69 %and 6.84 % thanMaximum likelihood and BP neural network, respectively.Meanwhile, the result also confirmed the practicability of CBERS-1 image in land use survey.
[1] Anderson J R, et al.A Land Use and Land Cover Classification System for Use with Remote Sensor Data.U S Geological Survey Professional Paper.1976 。964
[2] Tucker C J, Townshend J R G, Goff T E.African land-cover classification using satellite data.Science, 1985, 227:369 ～ 375
[3] Defries R S, Townshend J R G.NDVI-derived land cover classifi cation at global scales.International Journal of Remote Sensing, 1994b,15:3567 ～3586
[4] 国防科工委系统工程一司.中巴地球资源卫星数据应用评价.2000
[5] Heermann P D, Khazenie N.Classification of multispectral remote sensing data using a back propagation neural network.IEEE Trans Geosci Remote Sensing, 1992,(1):81 ～ 88
[6] Carpenter G A, Gjiaja M N, Gopal S, Woodcock C E.ART neural networks for remote sensing :Vegetation classif ication f rom Landsat TM and terrain data.IEEE Trans Geosci Remote Sens, 1997, 35(2):308 ～ 325
[7] 杨建刚。人工神经网络实用教程。杭州:浙江大学出版社, 2001
[8] 周志华, 等。自适应谐振理论综述。计算机科学,1999, 26(4):54 ～ 56
[9] 阎平凡, 张长水。人工神经网络与模拟进化计算。北京:清华大学出版社, 2000
[10] Fi tzgerald R W, Lee B G.Assessing the classificationAccuracy of multi source Remote Sensing dat a.Remote Sensing Environ, 1994, 47:362 ～ 368
/
| 〈 |
|
〉 |
