Automatic Interpretation of Chemical Structure Diagrams

摘要: Chemical structure diagrams, just as in engineering drawings, maps, and other technical consist of solid dashed lines (bonds), characters (atom symbols), symbols such brackets, parentheses, wedges (stereo-up bonds) or (stereo-down bonds). In addition to recognizing these low-level elements artifacts may be present — bonds intersections crossings atom nodes, character strings represent underlying chemical structure, circles are sometimes used ring-alternating bonding requiring a considerable knowledge base chemistry able interpret correctly. This paper discusses the general processes program Kekule 1 that embodies this interpretation ability with more detailed explanations how some problems relating polygon approximation, line wedge finding, optical recognition were solved.