A Constraint Network for Symbol Detection in Architectural Drawings
Résumé
A network used to detect and recognize several different symbols (doors, windows...) in scanned architectural drawings is presented. This network, based on Messmer's network for exact and inexact graph matching, presents a compact representation of all the symbols, which allows a one-pass search. Some modifications to this method for our specific document analysis problem are outlined: the symbols are represented as sets of constraints on the segments and arcs of the symbol; a description language has been written in order to describe these constraints. Symbol detection is performed by propagating the segments and the arcs in the network and retrieving the recognized symbols. The construction of this network is an iterative incremental process.