High School Statistical Graph Classification Using Hierarchical Model for Intelligent Mathematics Problem Solving

摘要

High school statistical graph classification is one of the key steps in intelligent mathematics problem solving system. In this paper, a hierarchial classification method is proposed for high school statistical graph classification. Firstly, the dense Scale-invariant Feature Transform (SIFT) features of the input images are extracted. Secondly, the sparse coding of the SIFT features are obtained. Thirdly, these sparse features are pooled in multiscale. Finally, these pooled features are concatenated and then fed into single-hidden layer feedforward neural network for classification. The effectiveness of the proposed method is demonstrated on the constructed dataset, which contains 400 statistical graphs. In contrast to several state-of-the-art methods, the proposed method achieves better performance in terms of classification accuracy, especially when the size of the training samples is small.