Affiliation:
1. Budapest University of Technology and Economics, Budapest, Hungary
Abstract
The article presents a new method developed to increase the efficiency of the identification algorithm for historical inscriptions of unknown origin. The authors extracted topological properties of the symbols containing different script relics, and analyzed them by using statistical tools. The considered topological properties are circular loop, oblique lines, vertical section, and crossing, among others. The article describes the use of the number of a three-circle grapheme intersection point vectors to identify unknown symbols. The number of intersections of the three circles and the examined symbol is stored in a feature vector. By supplementing the feature vectors with the circle vectors, the authors succeeded in improving the efficiency of the algorithm designed to decipher hard to read historical inscriptions.
Reference24 articles.
1. A review on handwritten character and numeral recognition for Roman, Arabic, Chinese and Indian scripts, International Journal of advanced studies in Computers;A. N.Azmi;Science & Engineering,2013
2. A novel feature extraction technique for the recognition of segmented handwritten characters, In;M.Blumenstein;Proceedings Seventh International Conference on,2003
3. HBF49 feature set: A first unified baseline for online symbol recognition
4. Encyclopedia of Distances
5. A structural/statistical feature based vector for handwritten character recognition
Cited by
4 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献