Imago: Open-Source Toolkit for 2D Chemical Structure Image Recognition

Different chemical databases contain molecule structures with links to articles and patents, where such molecules are presented as images. Keeping such a database in a consistent state is a challenging problem, and, thus, efficient and accurate molecule image recognition algorithms are very important for solving this task. We present an open-source toolkit for 2D chemical structure image recognition, called Imago. The main aim of this paper is to describe the algorithm approach implemented in Imago, and to share our ideas of possible improvements in the algorithm after we have summarized the results from the participation in the Image2Structure task at TRECCHEM 2011.