REMOTE SENSING IMAGE CLASSIFICATION BASED ON LOGISTIC MODEL
LIU Qing-sheng1, LIU Gao-huan1, LIN Qi-zhong2, WANG Zhi-gang2
1. State key Laboratory of Resources and Environmental information system, CAS, Beijing 100101, China;
2. Institute of Remote Sensing Applications, CAS, Beijing 100101, China
Logistic method is a nonlinear regression analysis method, which is based on the Logistic model. Usually, It is used to forecast and classify the unknown units into the known types. Other than the common classification methods, it can respectively calculate the probabilities which one unit belongs to the different known types, then, classify and forecast all the units of the unknown research field. In this paper, firstly we introduce the keystone of Logistic method, then, classify the two different remote sensing image data of the two different fields in The Inner Mongolia Autonomous Region by this method, finally discuss about a few factors which affect remote sensing image classification using logistic method.
刘庆生, 刘高焕, 蔺启忠, 王志刚 . 基于逻辑斯蒂模型的遥感图像分类[J]. 国土资源遥感, 2001, 13(1): 36-41,65.
LIU Qing-sheng, LIU Gao-huan, LIN Qi-zhong, WANG Zhi-gang . REMOTE SENSING IMAGE CLASSIFICATION BASED ON LOGISTIC MODEL. REMOTE SENSING FOR LAND & RESOURCES, 2001, 13(1): 36-41,65.
[1] 周宏坤.大比例尺矿床统计预测手册[S].北京:中国有色金属总公司地质研究所,1988.[2] Conese C et al. Improvement in Maximum Likelihood Classification performance on highly rugged using Principal Components Analysis【J】. INT.J.REMOTE SENSING, 1993, 14(7), 1371-1382.[3] 王志刚,朱振海.光谱角度填图在实验室光谱分析和岩性填图中的应用【J】.遥感学报,1999,3(1):60-65.[4] Paolo Gamba et al., A Completely Fuzzy Classification Chain for Multi-spectral Remote Sensing Images【J】. 1996,IEEE.[5] Yang H, van der Meer F, Bakker W, Tan Z. A back-propagation neural network for mineralogical mapping from AVIRIS data【C】. The third international airborne remote sensing conference and exhibition, 1997.