Aiming at the low accuracy and efficiency of art style painting classification algorithm, an art image attribute classification method based on vector machine(SVM) is proposed. Firstly, the scheme adopts the image classification architecture process based on SVM to fully extract the multi-dimensional features of the image and establish the feature database. Then the hierarchical classification method is used, and the specific SVM binary tree combined with the hybrid kernel classifier is adopted for image recognition and classification. Finally, illustrated by the attribute classification of ceramic art image, the experimental results show that the system has good recall and accuracy, which makes the image classification system based on SVM more widely used and more efficient.