The UvA color document dataset

Leon Todoran, Marcel Worring, Arnold W. M. Smeulders

Abstract: Publications on color document image analysis present results on small, nonpublicly available datasets. In this paper we propose a well-defined and groundtruthed color dataset consisting of over 1000 pages, with associated tools for evaluation. As we focus on aspects specific to color documents, we leave out the document textual content in the ground truth. The color data groundtruthing and evaluation tools are based on a well-defined document model, complexity measures to assess the inherent difficulty of analyzing a page, and well-founded evaluation measures. Together they form a suitable basis for evaluating diverse applications in color document analysis. Both the dataset and the tools are available through our Web site at http: //www.science.uva.nl/UvA-CDD