Article

Article title DEVELOPMENT OF HANDWRITING RECOGNITION ALGORITHM BASED ON RASTER GRAPHICS
Authors S.K. Bukhantseva
Section SECTION I. MATHEMATICAL METHODS OF SYNTHESIS OF SYSTEMS
Month, Year 02, 2013 @en
Index UDC 681.3.06
DOI
Abstract Flat images recognition methods using reference sequences were considered. The proposed recognition system is different from the previously used by approach that method of tracing the contour was used at the stage of segmentation for object bounds re-presentation, and the stage of character recognition is recognition by its skeleton. Using this approach can improve the performance of pattern recognition and can be used in the implementation of neural networks by changing its input parameters for signs of fonts. Furthermore, this system allows to add a variety of features that will increase the quality of pattern recognition.

Download PDF

Keywords Algorithm; bitmapped graphics; the reference sequence; segmentation; character recognition.
References 1. Sanjay Sengupta and Hyun S. Yang. Morphological Shape epresentation and Recognition of Binary Images // Proceedings of SPIE. – Jan. 1987. – Vol. 0848. – P. 51-56.
2. Рейер И., Петровцева М. Язык гранично-скелетного представления бинарных изображений // Доклады международной конференции Graphicon. – М., 2003. – C. 6.
3. Саймон Хайкин. Нейронные сети. Полный курс. – M.: Изд-во “Вильямс, 2006. – 1104 с.
4. Plamondon R., Srinari S. On-Line and Off-Line Handwriting Recognition: A Comprehensive Survey // IEEE Transactions on Pattern Analysis and Machine Intelligence. – January 2000.
– Vol. 22, № 1. – P. 4-20

Comments are closed.