Sub-pixel mapping method based on BP neural network

LI Jiao¹, WANG Li-guo², ZHANG Ye¹ , GU Yan-feng¹



Author NameAffiliation

LI Jiao Dept.of Information Engineering, Harbin Institute of Technology, Harbin 150001, China, wangliguo@hrbeu.edu.cn 

WANG Li-guo College of Information and Communications Engineering, Harbin Engineering University, Harbin 150001, China 

ZHANG Ye Dept.of Information Engineering, Harbin Institute of Technology, Harbin 150001, China, wangliguo@hrbeu.edu.cn 

GU Yan-feng Dept.of Information Engineering, Harbin Institute of Technology, Harbin 150001, China, wangliguo@hrbeu.edu.cn 



Abstract:

A new sub-pixel mapping method based on BP neural network is proposed in order to determine the spatial distribution of class components in each mixed pixel. The network was used to train a model that describes the relationship between spatial distribution of target components in mixed pixel and its neighboring information. Then the sub-pixel scaled target could be predicted by the trained model. In order to improve the performance of BP network, BP learning algorithm with momentum was employed. The experiments were conducted both on synthetic images and on hyperspectral imagery (HSI). The results prove that this method is capable of estimating land covers fairly accurately and has a great superiority over some other sub-pixel mapping methods in terms of computational complexity.

Key words:  sub-pixel mapping  BP neural network  BP learning algorithm with momentum

DOI：10.11916/j.issn.1005-9113.2009.02.027

Clc Number:TP273

Fund: