DeepChart: Combining deep convolutional networks and deep belief networks in chart classification

摘要

Chart classification is vital to chart analysis and document understanding. In this paper, we propose a novel framework (DeepChart) to classify charts by combining convolutional networks and deep belief networks. In general, we first extract deep hidden features of charts, which are taken from the fully-connected layer of deep convolutional networks. We then utilize deep belief networks to predict the labels of the charts based on their deep hidden features. The convolutional networks are initialized using a large number of natural images and fine-tuned using the chart images to avoid overfitting. Compared with previous methods using primitive feature extraction, the deep features achieve better scalability and stability. We collect a 5-class chart data set with more than 5000 images and demonstrate that the proposed framework greatly outperforms existing methods.