Handwritten Bangla City Name Recognition Using Shape-Context Feature
A segmentation-free approach is proposed to recognize the handwritten city names written in Bangla script. Initially, all the word images are converted into virtually single connected component following the refraction properties of light in order to design a unique shape-context of the same. Then a 64-dimensional feature vector is estimated from the said shape-context of each word image. A database containing 150 samples of 50 most popular city names of West Bengal, a state of India is prepared for evaluating the present method. Performance of this feature vector is also compared with some recently published feature vectors, and it is observed that the newly designed feature vector has outperformed the others.
KeywordsShape-context feature Handwritten word recognition City name recognition Bangla script
- 3.Sarkar, R., Malakar, S., Das, N., Basu, S., Kundu, M., Nasipuri, M.: Word extraction and character segmentation from text lines of unconstrained handwritten Bangla document images. J. Intell. Syst. 20(3), 227–260 (2011)Google Scholar
- 4.Bhowmik, S., Roushan, M.G., Polley, S., Malakar, S., Sarkar, R., Nasipuri, M.: Handwritten Bangla word recognition using HOG descriptor. In: Fourth International Conference of Emerging Applications of Information Technology (EAIT), pp. 193–197. IEEE (2014)Google Scholar
- 5.Bhowmik, S., Malakar, S., Sarkar, R., Nasipuri, M.: Handwritten Bangla word recognition using elliptical features. In: 2014 International Conference on Computational Intelligence and Communication Networks (CICN), pp. 257–261. IEEE (2014)Google Scholar
- 8.Singh, B., Mittal, A., Ansari, M.A., Ghosh, D.: Handwritten Devanagari word recognition: a curvelet transform based approach. Int. J. Comput. Sci. Eng. 3(4), 1658–1665 (2011)Google Scholar
- 10.Lavrenko, V., Rath, T.M., Manmatha, R.: Holistic word recognition for handwritten historical documents. In: Proceedings of First International Workshop on Document Image Analysis for Libraries, pp. 278–287. IEEE (2004)Google Scholar
- 11.Acharyya, A., Rakshit, S., Sarkar, R., Basu, S., Nasipuri, M.: Handwritten word recognition using MLP based classifier: a holistic approach. Int. J. Comput. Sci. Issues 10(2), 422–427 (2013)Google Scholar
- 13.Languages with at least 50 million first-language speakers. https://www.ethnologue.com/statistics/size. Accessed from Summary by language size Ethnologue
- 14.Das, B., Bhowmik, S., Saha, A., Sarkar, R.: An adaptive foreground-background separation method for effective binarization of document images. In: proceedings on 8th International Conference on Soft Computing and Pattern Recognition (2016)Google Scholar
- 15.Refraction. https://en.wikipedia.org/wiki/Refraction.