In recent years, with the development of the theory about Artificial Neural Network(ANN) system, the neural network technology is becoming increasingly an effective means of classification processing of remote senser digital images and beginning to replace the Maximum Likelihood Classifier(MLC). This paper discusses some ANN methods that have been very effectively applied abroad. these methods have vast prospects on digital image processing. At last, some principled problems about the practical application of the methods to remote sensing data classification are discussed.
张宝光. 人工神经网络在遥感数字图像分类处理中的应用[J]. 国土资源遥感, 1998, 10(1): 21-27.
Zhang Baoguang . THE APPLICATION OF ARTIFICIAL NEURAL NETWORK TO CLASSIFICATION PROCESSING OF REMOTE SENSING DIGITAL IMAGES. REMOTE SENSING FOR LAND & RESOURCES, 1998, 10(1): 21-27.
[1] Zhang Baoguang. Neural Networks and Fariergency Decision. 28th International Geographical Congmss. 5-10, August 1996. (aaepted for publicati} in Proc IGC96)[2] Mather P M. Computer Processing of Remotely Sensed Images. An Introduction, 1989[3] Lippman R P. An Introducriai to Comput吨with Neural Nets. IEEE Awustic Speech and Signal Processing Magazine.4-22, April 1987[4] Federiao G, et al. Representation Properties of Networks: Kolmogorov's Theorem is Irrelevant. Neural Computation,1989, Vol.l[5] Solaiman B, et al. A Canperative Study of Com}entional and Neural Network Classification讨Multispectral Data.Proceeding of IGARSS' 94, 1994, Vol III,1413-1415[6] Dawson M, et al. Neural Networks and their Applications to Parameter Retrieval and G. Newsletter. IEEE Geaeci. Remote Sensing Society, 1993, 6-14[7] Chen MSet al. Power Series Analysis of Back Propagation Neural Networks. Proc TCNN, 1991, 295-300[8] Kohonen T. Self一Chganizatiai and Associative Memory. 1989[9] Solaiman MQet al. A Hybrid Algorithm. HLVQ, Conbining Unsupervised and Supervised Ixaming Approaches.International Conferences on Neural Networks. ICNN 94, 1994
1998, Vol. 10 
