基于模煳小脑模型神经网络的遥感图像分类算法.pdf

31 4 Vol . 31, No. 4 2002 11 ACTA GEODAETICA et CARTOGRAPHICA SINICA Nov. , 2002 : 1001- 1595( 2002 04- 0327- 06 :TP753 : A 1, 2 1, 2 1 毛建旭 , 王耀南 , 孙 炜 ( 1. , 410082; 2. , 100080 Remote Sensing Image Classification Algorithm Based on Fuzzy CMAC Neural Network 1, 2 1, 2 1 MAO Jian-xu , WANG Yao- nan , SUN Wei (1. College of Electrical and Inf ormation Engineering , Hunan University, Changsha 410082, China; 2. NationalLaborato- ry of attern Recognition, CAS, Beijing 100080, China) Abstract: Considering the features of remote sensing images, w e proposed a remote sensing image classif ication algorithm using Fuzzy Cerebellar Model Articulation Controller ( FCMAC neural network . First , the principle of Cerebellar Model Ar- ticulation Controller( CMAC neural netw ork is described . Then, fuzzy theory is introduced into CMAC and a FCMAC neural network is brought forward. The proposed FCMAC neural network reflects the fuzziness and continuity of human cerebella and is used to remote sensing image classification. Experimental results show that the FCMAC neural network classifier can be used in remote sensing image classif ication , and its classif ication precision is superior to that of the conventional MLC al- gorithm . Key words: Cerebellar Model Articulation Controller; Fuzzy Cerebellar Model Articulation Controller; neural netw ork ; re- mote sensing image classif ication : 针对遥感图像分类的特点, 提出一种基于模糊小脑模型神经网络的遥感图像分类算 法首先阐 小脑模型神经网络的工作原理, 然后将模糊理论引入小脑模型神经网络, 提出一