Vision based of tactile paving detection method in navigation system for blind person

• Main Authors: Mohamed Kassim, Anuar, Yasuno, Takashi, Mohd Aras, Mohd Shahrieel, Shukor, Ahmad Zaki, Jaafar, Hazriq Izzuan, Baharom, Mohamad Faizal, Jafar, Fairul Azni
• Format: Article
• Language: English
• Published: Penerbit UTM Press 2015
• Subjects: T Technology (General)
• Online Access: http://eprints.utem.edu.my/id/eprint/18910/2/Journal-Co-JT2015an2a.pdf; http://eprints.utem.edu.my/id/eprint/18910/; http://www.jurnalteknologi.utm.my/index.php/jurnalteknologi/article/view/6547; http://dx.doi.org/10.11113/jt.v77.6547
• Institution: Universiti Teknikal Malaysia Melaka

In general, a blind person is relying on guide canes in order to walk outside besides depending only on a tactile pavement as warning and directional tools in order for avoiding any obstructions or hazardous situation. However, there still need a lot of training in order to recognize the tactile pattern and it’s quite difficult for new blind person. This paper describes the development and evaluation of vision based on the tactile paving detection method for blind person. The developed tactile paving detection system can be used by blind person for blind navigation purpose. Experiments will be conducted on how it works to detect the tactile pavement. In this experiment, we propose the vision based method by using MATLAB including Arduino platform and speaker as warning tools. There are two types of tactile pattern need to recognize which are circle acts as warning tactile and bar acts as directional tactile. Output of this system will be based on the result found from the tactile detection in MATLAB, then produces auditory output and notifies the blind about the type of tactile detected. As a result of this project, it will definitely help the blind by easing their mobility.