Technology and Methodology |
|
|
|
|
|
A METHOD FOR REMOTE SENSING IMAGE CLASSIFICATION BASED ON WEIGHT AND MIXED-PIXEL MODELS |
HE Hai-qing 1,2,LI Fa-bin 3 ,LI He-chao 3 ,WANG Yong 1,2 |
1. Graduate University of Chinese Academy of Sciences,Beijing 100049,China;2.Institute of Mountain Hazards and Environment,CAS,Chengdu 610041,China;3.Sichuan Research Institute of Territorial Surveying and Planning,Chengdu 610031,China |
|
|
Abstract A method for remote sensing image classification based on weight and mixed-pixel models is proposed in this paper. Combined with the available spectral mixed-pixel models,this method classifies the pixels by weighting and averaging with the abundances and weight factors of the categories according to practical application. To verify its feasibility,the authors carried out the classification of land cover with SPOT-5 data. The results show that the method can improve the accuracy of remote sensing image classification and even has higher application value under certain conditions.
|
Keywords
Artificial neural network
Remote sensing digital images
Classification
|
|
Issue Date: 15 July 2009
|
|
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|