Optical Braille Recognition Based on Histogram of Oriented Gradient Features and Support-Vector Machine

Abstract

In this paper, Braille character recognition in an image of a single side Braille document based on support vector machine (SVM) with Histogram of Oriented Gradient (HOG) as the feature extraction method is discussed. The developed system consists of two main stages, such as Braille character segmentation based on image processing techniques and Braille character recognition using SVM with HOG feature extraction method. The identified Braille characters can be translated to Sinhala language. In addition to that the developed system capable of identifying and translate to Grade 1 English braille characters and some of Grade 2 English Braille characters. The optical Braille character translating system was developed in MATLAB environment and it has proved more than 80% accuracy in most cases for translating Braille characters to either Sinhala or English.