UTHM Institutional Repository

Robust braille recognition system using image preprocessing and feature extraction algorithms

Taha, Hind Mowafaq (2014) Robust braille recognition system using image preprocessing and feature extraction algorithms. Masters thesis, Universiti Tun Hussein Onn Malaysia.


Download (1MB)


Braille Character Recognition (BCR) is a method to locate and recognize Braille document stored as an image, such as a jpeg, jpg, tiff or a gif image, and convert the text into a coded machine form such as text file. BCR converts the pixel representation of an image into its equivalent character representation. Braille recognition has plenty of benefits which facilitate the work in our daily life as workers in visually impaired schools and institutes. Based on literature review studies and remarks it can be concluded that extracting information fiom braille paper requires accuracy in pre-processing stage. Another approach is the reconlrnendation of matching and feature extraction that requier to be enhanced for optimal detection. This work is tested with a variety of Braille documents written using English Braille standards. The applied algorithm based on a comparison of Braille character location extraction in each cell with the templates created for each Braille cell. Many digital image processing stages have been implemented on the Braille document that can be imported to the system using scanner or camera, Gray scale conversion, binary conversion, filtering and inorphological dilation have been applied in the preprocessing stage which result in enhanced quality of Braille dots. Furthermore, edge detection, image projection and image segmentation of Braille document applied will improve the matching method. The proposed method in this project succeeds effectively extract Braille dots fiom the paper Braille picture, and compare it with the provided templates of characters and then transform it to English text file. This project development is based on MATLAB 201 1 software programming. The implemented algorithm achieved 100% precise results where several cases have been performed with excellent recognition outcomes.

Item Type: Thesis (Masters)
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA1501-1820 Applied optics. Photonics
Depositing User: Normajihan Abd. Rahman
Date Deposited: 01 Jul 2014 02:19
Last Modified: 01 Jul 2014 02:19
URI: http://eprints.uthm.edu.my/id/eprint/5519
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item


Downloads per month over past year