A new algorithm for image recognition and classification based on improved Bag of Features algorithm

Abstract

This paper presents a method for image recognition and classification based on improved Bag of Features (BOF). In view of the low efficiency and low classification accuracy of the traditional BOF algorithm, a new recognition and classification algorithm combined speeded-up robust features (SURF) and spatial pyramid matching principle is proposed in this paper. SURF algorithm can improve the efficiency and spatial pyramid matching principle can improve the classification accuracy. This paper's method uses SURF algorithm to extract the image feature and generate the codebook. The spatial pyramid matching principle is applied to the image histogram's codebook which can improve the accuracy of the classification. Finally, the method uses the LIBSVM classifier to classify the image histogram's codebook. The experiments are carried out based on Graz, Caltech-256, and Pascal VOC 2012. The results show that our proposed method is better than the traditional method in the efficiency and classification accuracy. In addition, our method is compared with some related research work in classification accuracy, and the results show that our algorithm has obvious advantages.