A Support Vector Machine Model for Mapping of Lake Water Quality from Remote-Sensed Images

Abstract

Abstract In this paper, a support vector machine (SVM) model is proposed for the mapping of water quality of lakes from remote-sensed images. Based on the remote-sensed image characteristics of Lake Tai in China, a set of feasible parameters are given for the SVM model after computer optimization. The distribution of the chlorophyll density of Lake Tai can be easily discovered from the proposed SVM model. Compared with the conventional linear regression method and the belief propagation (BP) neural network method, the proposed SVM model can outperform the traditional ones at both monitored points and non-monitored points.