An evolutionary algorithm for general symbol segmentation

A new system is presented for generalsymbol segmentation, which is applicable forsegmentation of any connected string of symbols,including characters and line diagrams. Using apowerful graph representation and anevolutionary algorithm framework, segmentationhypotheses are initialized and evolved towards afully segmented and recognized string. Theevolutionary segmentation was tested in manydomains including connected digits, connectedcharacters and simple circuit diagrams. Theperformance of the evolutionary algorithmdepends heavily on the symbol recognitionsystem used.

[1]  Marc Parizeau,et al.  Learning to Segment Cursive Words using Isolated Characters , 1999 .

[2]  Proceedings Seventh International Conference on Document Analysis and Recognition , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[3]  Rabab Kreidieh Ward,et al.  A Rotation Invariant Rule-Based Thinning Algorithm for Character Recognition , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Ching Y. Suen,et al.  A new system for reading handwritten zip codes , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[5]  Ashok Samal,et al.  A system for recognizing a large class of engineering drawings , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Rabab Kreidieh Ward,et al.  An expert system for general symbol recognition , 2000, Pattern Recognit..

[7]  Ashok Samal,et al.  A system for recognizing a large class of engineering drawings , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[8]  Réjean Plamondon,et al.  A handwriting model for syntactic recognition of cursive script , 1992, Proceedings., 11th IAPR International Conference on Pattern Recognition. Vol.II. Conference B: Pattern Recognition Methodology and Systems.