Accuracy in handwritten character recognition system is a challenge in the area of pattern recognition because of a variety of writing styles. Eigenface is a method that has been widely used in face recognition systems. This method is proposed in the field of handwritten character recognition, in this paper. Here, Eigencharacters are created from a 2-D training set of images and weight vectors are generated. These weight vectors are used as feature vectors for classification. The classification is performed using Euclidean Distance, k-NN and SVM classifiers. Experimental results proved that the proposed Eigencharacter method using Euclidean distance produced good classification accuracy.
CITATION STYLE
Ashlin Deepa, R. N., & Rajeswara Rao, R. (2017). An eigencharacter technique for offline-tamil handwritten character recognition. In Advances in Intelligent Systems and Computing (Vol. 458, pp. 495–505). Springer Verlag. https://doi.org/10.1007/978-981-10-2035-3_51
Mendeley helps you to discover research relevant for your work.