抽象的な

An improved image classification method basd on multi features using fuzzy support vector machine

Cao Jianfang, Chen Junjie, Chen Lichao

The development of electronic technology and multimedia technology has led to the rapid growth of digital images. It has become an important problem to be solved to depend on advanced technology to classify images. We proposed an image classification method based on multi features using fuzzy support vector machine. It extractes various features and uses fuzzy support vector machine to classify images. The proposed method overcomes the shortcoming of traditional support vector machine in multi-classification problems and solves the problem of semantic ambiguity in image classification field by using fuzzy membership function. Using 6 types of natural images to train and test, the experiment results show that classification performance improves significantly compared with the traditional support vector machine algorithm. It lays a good foundation for further improving digital image understanding.