Research and Design of Image Feature Recognition Classifier Based on SVM

It built a classifier, which could improve recognition rate of image feature. Based on statistical learning theory, it built a classifier on support vector machine (SVM), and determined the parameter of SVM and Guass radial kernel function. In the experiment, classifier of SVM was trained by feature sample, then carried on classifying , recognition and detection. The result of simulation showed that classification based on support vector machine (SVM) not only has better robustness, but also effectively improve the recognition rate and decrease false recognition rate .