• M.Sivasankari

Manuscripts transmit messages, ideas and thoughts in writing with documentary proof which is a form of communication. This paper majorly focuses on the Feature extraction that is the process to retrieve the characters that are represented accurately from the raw data. The purpose of feature extraction is to maximize the recognition rate with the characters. It is considered to be a difficult task to obtain the features of the handwriting which has high degree of variability and imprecision. In this paper the steps namely Denoising ,binarization, noise elimination, Thresholding, and size normalization for an image are carried out then words segmentation is performed by using bat algorithm. In feature extraction stage each character’s identity is represented as a feature vector. In feature extraction the set of words have been performed through structural and statistical methods. To identify the style of words and characters, fuzzy clustering algorithm has been used. The efficient recognition of word has been done through the online support vector classifier. The result shows that the proposed methods have been performed well when compared with the existing methods.

Keywords: Off-Line Handwriting Recognition, Image Denoising, Binarization, Noise Elimination, Size Normalization, Bat Algorithm, Edge Detection, Fuzzy Clustering, and Online Support Vector Machine (OSVM).

"MULTILINGUAL OFF-LINE HANDWRITING RECOGNITION IN REAL-WORLD IMAGES USING THE ONLINE SUPPORT VECTOR MACHINE", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 4, page no.603-618, April-2019, Available :http://www.jetir.org/papers/JETIR1904E89.pdf

"MULTILINGUAL OFF-LINE HANDWRITING RECOGNITION IN REAL-WORLD IMAGES USING THE ONLINE SUPPORT VECTOR MACHINE", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 4, page no. pp603-618, April-2019, Available at : http://www.jetir.org/papers/JETIR1904E89.pdf