Model-Based Chart Image Recognition

In this paper, we introduce a system that aims at recognizing chart images using a model-based approach. First of all, basic chart models are designed for four different chart types based on their characteristics. In a chart model, basic object features and constraints between objects are defined. During the chart recognition, there are two levels of matching: feature level matching to locate basic objects and object level matching to fit in an existing chart model. After the type of a chart is determined, the next step is to do data interpretation and recover the electronic form of the chart image by examining the object attributes. For testing purpose, a set of testing images are either downloaded from the internet or scanned from books and papers. By comparing the recovered data and the original chart, we are able to evaluate the current system and confirm its validity.