Title :
Symbol Spotting in Line Drawings through Graph Paths Hashing
Author :
Dutta, Anjan ; Lladós, Josep ; Pal, Umapada
Author_Institution :
Comput. Vision Centre, Univ. Autonoma de Barcelona, Barcelona, Spain
Abstract :
In this paper we propose a symbol spotting technique through hashing the shape descriptors of graph paths (Hamiltonian paths). Complex graphical structures in line drawings can be efficiently represented by graphs, which ease the accurate localization of the model symbol. Graph paths are the factorized substructures of graphs which enable robust recognition even in the presence of noise and distortion. In our framework, the entire database of the graphical documents is indexed in hash tables by the locality sensitive hashing (LSH) of shape descriptors of the paths. The hashing data structure aims to execute an approximate k-NN search in a sub-linear time. The spotting method is formulated by a spatial voting scheme to the list of locations of the paths that are decided during the hash table lookup process. We perform detailed experiments with various dataset of line drawings and the results demonstrate the effectiveness and efficiency of the technique.
Keywords :
object detection; object recognition; Hamiltonian path; complex graphical structures; entire database; graph paths hashing; hash table lookup process; hashing data structure; k-NN search; line drawing; model symbol; robust recognition; shape descriptor; spatial voting scheme; spotting method; symbol spotting; Databases; Noise; Shape; Table lookup; Vectors; Graph factorization; Graph paths hashing; Graphics recognition; Shape descriptors; Symbol spotting;
Conference_Titel :
Document Analysis and Recognition (ICDAR), 2011 International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4577-1350-7
Electronic_ISBN :
1520-5363
DOI :
10.1109/ICDAR.2011.199