Uncertainty Research of Remote Sensing Image Classification based on Hybrid Entropy Evaluation Model
This study put forward an integrated evaluation model.Bases on a framework of fuzzy set theory and entropy theory,we firstly complete the classification using fuzzy surveillance approach,taking it as a formalized description of classification uncertainty.Then introduce hybrid entropy model for classification uncertainty evaluation,which can meet the requirement of comprehensive reflection of both random and fuzzy uncertainty,meanwhile construct evaluation index from pixel scale with the full consideration of different contribution to error rate of each pixel.Finally,we use such method to evaluate land-use classification result of remote sensing image,which is in Huangshi city,Hubei province of China,by using hybrid entropy evaluation model,the classification quality can be fully reflected,and pixelscale evaluation indexes were easier constructed.
uncertainty of RS image classification fuzzy surveillance classification hybrid entropy evaluation model pixel-scale evaluation indexes
Zeying Lan Yanfang Liu Xiangyun Tang Gang Liu
School of Resource and Environment Science,Wuhan University,Wuhan 430079,China;Key Laboratory of Geo School of Resource and Environment Science,Wuhan University,Wuhan 430079,China;Key Laboratory of Geo School of Electronic and informat,Wuhan University,Wuhan 430079,China
国际会议
第16届国际地理信息科学与技术大会(16th International Conference on GeoInformatics and the Joint Conference)
广州
英文
2008-06-28（万方平台首次上网日期，不代表论文的发表时间）