Bangla Character Recognition using Artificial Neural Network Step:Classifier Design

Abstract

Character recognition is a very popular research field since 1950’s. A great deal of research work has been done for various languages specifically in case of English. The recognition of optical characters is known to be one of the earliest applications of Artificial Neural Networks. The use of artificial neural network simplifies development of an optical character recognition application, while achieving highest quality of recognition and good performance. Although Bangla is one of the most widely spoken languages (over 200 million people use Bangla as their medium of Communication) of the world, research is acute in recognition of Bangla characters. Under this context, an effort has been taken globally to computerize the Bangla language. Compared to English and other language scripts, one of the major stumbling blocks in Optical Character Recognition (OCR) of Bangla script is the large number of complex shaped character classes of Bangla alphabet. In addition to 50 basic character classes, there are nearly 160 complex shaped compound character classes in Bangla alphabet. Dealing with such a large variety of characters with a suitably designed feature set is a challenging problem. Uncertainty and imprecision is inherent in handwritten script. Moreover, such a large variety of complex shaped characters, some of which have close resemblance, makes the problem of OCR of Bangla characters more difficult. Considering the complexity of the problem, this research makes an attempt to develop a method for the recognition of Bangla characters using the artificial neural network. Pre-processing steps involves segmentation and binarization. Features are taken using different feature extraction procedures. Multilayered neural network is used in the spirit of back-propagation algorithm for classification as well as recognition of characters. To deal with immense variation and magnificent diversity of Bangla characters, this effort have widened the area for many research works to come to light and to bid fair to be accomplished.