Constructing a Hierarchical Structure from Symbol Alphabets of Technical Line Drawings

This paper presents a method for analysing symbol alphabets of technical line drawings and finding their underlying structure, which is important for investigating (dis)similarity of different symbols. The proposed method constructs a hierarchical structure of a set of technical symbols. The method is based on agglomerative hierarchical clustering that uses either of two variants as a similarity measure: either geometric matching between symbols' shapes, or an off-the-shelf shape descriptor. Identifying such a hierarchical structure of a set of symbols can improve symbol recognition / spotting systems, as it helps with scalability issues, and provides information on the degree of similarity among symbols, so that those systems can automatically adapt their parameter values for more accurate recognition. Our method has been tested on the symbol alphabet of the symbol recognition / spotting contest of GREC-2011, and achieved promising results.

[1]  Jean-Yves Ramel,et al.  Graphic Symbol Recognition Using Graph Based Signature and Bayesian Network Classifier , 2009, 2009 10th International Conference on Document Analysis and Recognition.

[2]  Thomas M. Breuel,et al.  On the Use of Geometric Matching for Both: Isolated Symbol Recognition and Symbol Spotting , 2011, GREC.

[3]  Liu Wenyin,et al.  A New Vectorial Signature for Quick Symbol Indexing, Filtering and Recognition , 2007 .

[4]  Alexander Wong,et al.  Robust Hough-Based Symbol Recognition Using Knowledge-Based Hierarchical Neural Networks , 2008, IPCV.

[5]  Kun Zhang,et al.  Symbol Recognition with Kernel Density Matching , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Salvatore Tabbone,et al.  Symbol Descriptor Based on Shape Context and Vector Model of Information Retrieval , 2008, 2008 The Eighth IAPR International Workshop on Document Analysis Systems.

[7]  Tong Lu,et al.  Symbol Recognition Combining Vectorial and Pixel-Level Features for Line Drawings , 2010, 2010 20th International Conference on Pattern Recognition.

[8]  Philip S. Yu,et al.  Rotation invariant indexing of shapes and line drawings , 2005, CIKM '05.

[9]  R. Sokal,et al.  Numerical Taxonomy: The Principles and Practice of Numerical Classification. , 1975 .

[10]  Josep Lladós,et al.  Hierarchical Graph Representation for Symbol Spotting in Graphical Document Images , 2012, SSPR/SPR.

[11]  Pasi Fränti,et al.  Content-based matching of line-drawing images using the Hough transform , 2000, International Journal on Document Analysis and Recognition.

[12]  Karell Bertet,et al.  Symbol Recognition Using a Concept Lattice of Graphical Patterns , 2009, GREC.

[13]  Min Feng,et al.  Symbol Recognition Using Bipartite Transformation Distance and Angular Distribution Alignment , 2005, GREC.

[14]  Anil K. Jain,et al.  Data clustering: a review , 1999, CSUR.

[15]  Mickaël Coustaty,et al.  On the Joint Use of a Structural Signature and a Galois Lattice Classifier for Symbol Recognition , 2007, GREC.

[16]  S. C. Johnson.  Hierarchical clustering schemes , 1967, Psychometrika.

[17]  Alexander Wong,et al.  Robust Invariant Descriptor for Symbol-Based Image Recognition and Retrieval , 2007, International Conference on Semantic Computing (ICSC 2007).

[18]  Su Yang.  Symbol Recognition via Statistical Integration of Pixel-Level Constraint Histograms: A New Descriptor , 2005, IEEE Trans. Pattern Anal. Mach. Intell..

[19]  Alexander Wong,et al.  Robust Invariant Descriptor for Symbol-Based Image Recognition and Retrieval , 2007 .

[20]  Jean-Yves Ramel,et al.  A Content Spotting System for Line Drawing Graphic Document Images , 2010, 2010 20th International Conference on Pattern Recognition.