Real-Time Canny-Edge Based Obstacle Detection Model for Mobile Devices with Monocular Camera

There are at least 253 million people in the world who are visually impaired. This thesis proposed a model of supplementary assistance for the people with low vision to detect obstacles in the walking path. The model is developed to be installed in a mobile device such as smart phone which eliminates the issues in mass-production and distribution of assistive devices. It uses a mobile device's camera and applies the edge-based obstacle detection algorithm in real-time. In addition, the secondary internal sensors such as the gyroscope and more are used to enhance the detection capability.