Scientific chart image recognition and interpretation

Abstract

This dissertation presents the research work on scientific chart image recognition and interpretation, a relatively new area of document image analysis. Literature review is conducted to summarize relevant research activities and point out their limitations that are to be overcome. In this dissertation, we provide a general chart recognition and interpretation paradigm, and investigate all the major aspects of the research problem, including chart image recognition that focuses on extracting low-level graphical symbols and text symbols for classification and chart component construction, chart interpretation that performs high-level association of textual and graphical information to capture the semantic meaning of chart images and generate descriptions, the application of the proposed methods including optical character recognition (OCR) and question answering (QA), and ground truth dataset generation for performance evaluation.