CAO Bao,QIN Qi-ming,ZHANG Zi-li,et al.Feature Enhancement Techniques Combined with Object-Oriented Classification Approach[J].Research of Soil and Water Conservation,2008,15(01):135-138.
基于特征增强技术的面向对象分类方法
- Title:
- Feature Enhancement Techniques Combined with Object-Oriented Classification Approach
- Keywords:
- feature enhancement; object-oriented; classification
- 分类号:
- TP391.4
- 摘要:
- 针对传统基于像元的分类方法不能满足对高分辨率影像（HRI）进行分类的矛盾，提出基于特征增强技术的面向对象分类方法（FETCOOCA）。并以北京市海淀区的SPOT5影像为例，考虑到影像中植被、水体和建筑物等地物之间的特征差异，对影像中地物的光谱、形状、纹理等信息进行特征增强处理，并结合面向对象的分类方法对其进行分类。最后，对FETCOOCA与传统基于像元的分类方法进行对比分析。结果表明：提出的FETCOOCA明显优于传统分类方法，它可以大幅度提高HRI的分类精度，有效抑制椒盐现象的发生，使分类后的图像含有更为丰富的语义信息。
- Abstract:
- To deal with the contradiction of classical pixel based on classification approaches cannot satisfy the requirements of High Resolution Image (HRI) classification.Feature Enhancement Techniques Combined with Object-Oriented Classification Approach (FETCOOCA) is provided.Considering the difference of the objects in terms of spectrum, shape, texture etc.contained in HRI, some feature enhancement techniques are carried out on the area of water body, vegetation, and constructions.To test the overall accuracy of FETCOOCA classification result image, FETCOOCA is performed using SPOT5 image, which located in Haidian District, Beijing, China.Finally, a comparison between FETCOOCA and classical pixel based classification approaches is carried out.The study proved that: for HRI classification, FETCOOCA has much better classification result than classical pixel based classification approaches.It can significantly improve the overall classification accuracy of HRI, can avoid the pepper and salt phenomena, and have rich syntax information which is easy for image interpretation.
参考文献/References:
[1] Skidmore A, Turner B, Brinkhof W, et al.Performance of neural network:mapping forests using GIS and remotely sensed data[J].ISPRS Journal of Photogrammetric Engineering & Remote Sensing, 1997, 63:501-514.
[2] 秦其明, 陆荣建.分形与神经网络方法在卫星数字图像分类中的应用[J].北京大学学报:自然科学版, 2000, 36(6):858-864.
[3] 周成虎, 骆剑承, 杨存建, 等.遥感图像地学理解与分析[M].北京:科学出版社, 2001.
[4] Maselli F, Rudolf A, Conese C.Fuzzy Classification of spatially degraded thematic mapper data for the estimation of sub-pixel components[J].International Journal of Remote Sensing, 1996, 17:537-551.
[5] 骆剑承, 梁怡, 周成虎.基于尺度空间的分层聚类方法及其在遥感图像分类中的作用[J].测绘学报, 1999, 28(4):319-324.
[6] Lobo A, Chic O, Casterad A.Classification of mediterranean crops with multisensor data:per-pixel versus per-object statistics and image segmentation[J].International Journal of Remote Sensing, 1996, 17(12):2385-2400.
[7] Aplin P, Atkinson P, Curran P.Per-field classification of land use using the forthcoming very fine resolution satellite sensors:problems and potential solutions[M]//Advances in Remote Sensing and GIS Analysis.Chichester:Wiley & Sons, 1999:219-239.
[8] Blaschke T, Lang S, Lorup E, et al.Object-oriented image processing in an integrated GIS/remote sensing environment and perspectives for environmental applications[C]//Environmental Information for Planning.Metropilis-Verlag, 2000, 2:555-570.
[9] BAATZM A SCHÄPE:Object-Oriented and Multi-Scale Image Analysis in Semantic Networks[C]//Proc.of the 2nd International Symposium on Operationalization of Remote Sensing.August 16th-20th 1999.Enschede.ITC.1999.
[10] BAATZ M A SCHÄPE:Multiresolution Segmentation-an optimization approach for high quality multi-scale image segmentation[C]//Strobl, Blaschke & Greisebener:Angewandte Geographische Informationsverarbeitung XI.Beit räge zum AGIT-Symposium Salzburg 1999.Karlsruhe.Herbert Wichmann Verlag.2000.
相似文献/References:
[1]刘英敏,杨天行,刘晓端.密云水库及上游流域环境监测信息系统的设计与实现[J].水土保持研究,2006,13(01):197.
LIU Ying-min,YANG Tian-xing,LIU Xiao-duan.Design and Implementation of Environment Monitoring Information System for Miyun Reservior Upper Valley[J].Research of Soil and Water Conservation,2006,13(01):197.
[2]王永杰,苗丽,周新忠.基于面向对象的元数据库设计实现网络地理空间元数据管理系统[J].水土保持研究,2007,14(05):278.
WANG Yong-jie,MIAO Li,ZHOU Xin-zhong.Design and Realize the Net Management System for Geospatial Metadata Based on Object-Oriented Metadata Base[J].Research of Soil and Water Conservation,2007,14(01):278.
[3]郭鹏,徐丽萍,王玲,等.面向对象的玛纳斯河流域TM影像分类技术研究[J].水土保持研究,2013,20(03):120.
GUO Peng,XU Li-ping,WANG Lin,et al.Research for Object-Oriented TM Image Segmentation and Information Extraction Technology[J].Research of Soil and Water Conservation,2013,20(01):120.
备注/Memo
收稿日期:2006-10-31
基金项目:北京市自然基金项目（9062006）；空间信息集成与3S工程应用北京市重点实验室项目（SⅡBKL06-3-03）
作者简介:曹宝（1971- ），男，助理研究员，博士研究生，主要从事遥感与地理信息技术应用研究。