A COMPARATIVE STUDY OF PROBABILISTIC NEURAL NETWORK AND BP NETWORKS FOR REMOTE SENSING IMAGE CLASSIFICATION
LI Chao-feng1, YANG Mao-long2, XU Lei1, YANG Meng-zhao1
1. School of Information Technology, Southern Yangtze University, Wuxi 214036, China;
2. Department of Space Reconnaissance, University of Foreign Relations, PLA, Nanjing 210039, China
This paper has analyzed the basic theory and algorithm of the probabilistic neural network, and established the remote sensing image classification model based on the probabilistic neural network. Examples show that the probabilistic neural network model outperforms the improved back-propagation neural network model in classification precision and is close to the latter in time consumption. It proves to be an efficient image classification method.
李朝锋, 杨茂龙, 许磊, 杨蒙召. 概率神经网络与BP网络模型在遥感图像分类中的对比研究[J]. 国土资源遥感, 2004, 16(4): 11-13，18.
LI Chao-feng, YANG Mao-long, XU Lei, YANG Meng-zhao . A COMPARATIVE STUDY OF PROBABILISTIC NEURAL NETWORK AND BP NETWORKS FOR REMOTE SENSING IMAGE CLASSIFICATION. REMOTE SENSING FOR LAND & RESOURCES, 2004, 16(4): 11-13，18.
[1] Atkinson P M, Tatnall A R L.Neural networks in remote sensing［J］.INT J Remote Sensing, 1997, 18(4):699-709.
[2] Bischof H, Schneider W, Pinz A.Multispectral classification of Landsat-images using neural networks［J］.IEEE Trans on Geoscience and Remote Sensing, 1992,30(3):482-490.
[7] Muvai H.Remote sensing image analysis using a neural network and knowledge-based processing［J］.International Journal of Remote Sensing,1997,18(4):811-828.
2004, Vol. 16 
