A HYBRID SCHEME FOR HANDPRINTED NUMERAL RECOGNITION BASED ON A SELF-ORGANIZING NETWORK AND MLP ClASSIFIERS
- 21 November 2002
- journal article
- Published by World Scientific Pub Co Pte Ltd in International Journal of Pattern Recognition and Artificial Intelligence
- Vol. 16 (7), 845-864
- https://doi.org/10.1142/s0218001402002027
Abstract
This paper proposes a novel approach to automatic recognition of handprinted Bangla (an Indian script) numerals. A modified Topology Adaptive Self-Organizing Neural Network is proposed to extract a vector skeleton from a binary numeral image. Simple heuristics are considered to prune artifacts, if any, in such a skeletal shape. Certain topological and structural features like loops, junctions, positions of terminal nodes, etc. are used along with a hierarchical tree classifier to classify handwritten numerals into smaller subgroups. Multilayer perceptron (MLP) networks are then employed to uniquely classify the numerals belonging to each subgroup. The system is trained using a sample data set of 1800 numerals and we have obtained 93.26% correct recognition rate and 1.71% rejection on a separate test set of another 7760 samples. In addition, a validation set consisting of 1440 samples has been used to determine the termination of the training algorithm of the MLP networks. The proposed scheme is sufficiently robust with respect to considerable object noise.Keywords
This publication has 18 references indexed in Scilit:
- Self-adaptive learning rates in backpropagation algorithm improve its function approximation performancePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Skeletonization by a topology-adaptive self-organizing neural networkPattern Recognition, 2001
- Off-line Arabic character recognition: the state of the artPattern Recognition, 1998
- A complete printed Bangla OCR systemPattern Recognition, 1998
- Neural network-based systems for handprint OCR applicationsIEEE Transactions on Image Processing, 1998
- Growing Grid — a self-organizing network with constant neighborhood range and adaptation strengthNeural Processing Letters, 1995
- Recognition of handwritten numerals with multiple feature and multistage classifierPattern Recognition, 1995
- Growing cell structures—A self-organizing network for unsupervised and supervised learningNeural Networks, 1994
- Self-creating and organizing neural networksIEEE Transactions on Neural Networks, 1994
- A simple and effective method for removal of hidden units and weightsNeurocomputing, 1994