Braille is a system that allows visually impaired people to write and read. It utilizes the finger touch on raised dots produced by specialized machine. In this paper we present automatic speech recognition (ASR) system which, is aimed at helping preschool children to learn Braille. In order to provide valuable support in decision making, a system combining Braille images and image mining becomes necessary. Such a system would encompass the following functions set of Braille images, image

more »

... cessing, character segmentation, feature extraction, and character recognition. Gray scale conversion, binary conversion, and complement have been applied in the pre-processing stage which helps extracts Braille dots. Centroid features and Bayesian classifier succeeds effectively to classify Braille to English alphabets. The implemented algorithm achieved 100% results where as classifiers such as Classification Tree and SVM could achieve only 23.0769 and 80.7692 respectively.