Abstract:

Aiming at the feature selection problem in image classification, we proposed a multiple feature selection and support vector machine for an image classification model. Firstly, we extracted a varity of features of the image and normalized the features. Secondly, we selected a set of optimal feature subset according to the average impact value of features. Finally, the multi classifier was built by using support vector machine, and the simulation experiments were carried out to verify the validity of the model by using image data set SIMPLIcity. The experimental results show that the proposed model can reduce the cost of image classification and improve the performance of image classification.

Key words: image processing, classification model, multiple feature, support vector machine