SVM-Based Remote Sensing Image Classification and Monitoring of Lijiang Chenghai

Support Vector Machine (SVM), which is based on the statistical learning theory, was used for remote sensing image classification. On the condition that ground investigation and artificial visual interpretation was used to distinguish features, the method used for classification accuracy assessment of the sample, which was chosen at random, was confusion matrix. The experimental results show that, in the condition of the same data set, the classification accuracy of SVM was significantly higher than that of the neural network and maximum likelihood classifier. Regarding water samples, which had been sampled at random, reached 100% classification accuracy based on SVM. Also because of such, the area of Chenghai was calculated to meet the demand of monitoring of the area of Chenghai.