Visual Impaired Person Navigation Assistance Using Motion Sensor

• Main Authors: Pebrianti, Dwi, Rosdiyana, Samad, Zainah, Md. Zain, Addie Irawan, Hashim, Mohd Razali, Daud
• Format: Conference or Workshop Item
• Language: English
• Published: 2013
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://umpir.ump.edu.my/id/eprint/5728/1/EE-005.pdf; http://umpir.ump.edu.my/id/eprint/5728/

According to the website of World Health Organization (WHO), there are about 285 million people are visually impaired worldwide. This can be separated into 39 million are blind and 246 have low vision. As the vision system has degraded, the person will have difficulty to navigate. In general, the visually impaired person will use a white cane or a seeing-eye dog. However, these systems are considered to be cumbersome due to the difficulties of the usage. For example, in some places, the seeing-eye dog cannot be brought inside, such as a restaurant. Another example is the white cane will have the possibility to be left somewhere. In order to solve the problems, a new technology is needed. In this research, a low cost motion sensor, Kinect sensor, will be used as a tool for assisting a visual impaired person to navigate in a complex environment, such as an indoor environment. The Kinect motion sensor, Figure 1 is completed with (from left to right) an IR projector, an RGB camera and an infrared camera. This device also has a four-microphone array in addition to a RGB camera. Therefore, it can provide depth signals, RGB images, and audio signals simultaneously.The configuration of the sensor on the Kinect sensor and the data captured from a Kinect sensor is shown in Figure 1.