Skip to Main Content
Support vector machine (SVM) is more popular in recent years in the field of pattern recognition algorithms. SVM algorithm has good validity of the calculation, statistical robustness and stability. SVM has gradually become an important tool in the field of remote sensing image classification. Because of the similarity between the different classes in the multispectral images, the result of the classification which is gotten by using the SVM directly is usually not satisfying. In this paper, we proposed a method based on fuzzy SVM (FSVM) to classify the multispectral images. The result of the experiment shows that the accuracy of this method is higher compared with the method which used the SVM directly.