On the recognition of Devanagari ancient handwritten characters using SIFT and Gabor features
- PDF / 1,813,416 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 8 Downloads / 220 Views
(0123456789().,-volV)(0123456789().,-volV)
METHODOLOGIES AND APPLICATION
On the recognition of Devanagari ancient handwritten characters using SIFT and Gabor features Sonika Rani Narang1
· M. K. Jindal2 · Shruti Ahuja1 · Munish Kumar3
© Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Recognition of Devanagari ancient handwritten character is an important task for resourceful contents’ exploitation of the priceless information contained in them. There are numerous Devanagari ancient handwritten documents from fifteenth to the nineteenth century. This paper presents an optical character recognition system for the recognition of Devanagari ancient manuscripts. In this paper, improved recognition results for Devanagari ancient characters have been presented using the scale-invariant feature transform (SIFT) and Gabor filter feature extraction techniques. Support vector machine (SVM) classifier is used for the classification task in this work. For experimental results, a database consisting of 5484 samples of Devanagari characters was collected from various ancient manuscripts placed in libraries and museums. SIFTand Gabor filter-based features are used to extract the properties of the handwritten Devanagari ancient characters for recognition. Principle component analysis is used to reduce the length of the feature vector for reducing training time of the model and to improve recognition accuracy. Recognition accuracy of 91.39% has been achieved using the proposed system based on tenfold cross-validation technique and poly-SVM classifier. Keywords Ancient manuscripts · Devanagari historical documents · SIFT · Gabor · SVM
1 Introduction Apart from diversified languages, India is a home to a remarkable array of scripts derived from the Brahmi script. These ancient scripts transmit knowledge about cultural and scientific heritage from one generation to another.
Communicated by V. Loia. & Munish Kumar [email protected] Sonika Rani Narang [email protected] M. K. Jindal [email protected] Shruti Ahuja [email protected] 1
Department of Computer Science, DAV College, Abohar, Punjab, India
2
Department of Computer Science and Applications, Panjab University Regional Centre, Muktsar, Punjab, India
3
Department of Computational Sciences, Maharaja Ranjit Singh Punjab Technical University, Bathinda, Punjab, India
These documents can be a valuable source of information, but they suffer progressive degradation over the time because of certain environmental factors, poor handling and ageing. The Devanagari script is one of the scripts, which are used in many Indian languages such as Hindi or Sanskrit. Devanagari had been used for writing various religious and urbane scriptures. Various museums and libraries have preserved a collection of ancient documents, but they are not easily accessible to the public. Because of their vital importance, these documents or their contents are needed to be preserved for future generations. As there is a strong move towards digitization, ma
Data Loading...
