Object Detection System to Help Navigating Visual Impairments

Abstract

The number of people with severe visual impairments and blind people in the world is 216.6 million and 38.5 million, respectively in 2018 and that number will increase every year. While the development of Computer Vision technology became popular after that method is used in automatic driving system using an object detection system to detect surrounding object, this technology can be a solution to help blind people too. This can be done by implementing Harris Corner Detection method. Harris Corner Detection method is used to detect the corner of the object in the image taken. The number of corner and location of corner in detection result can be used for predicting position and distance. To predict the distance, a triangle rule will be used in finding the distance. Furthermore, it can predict the location and distance of the object in the picture taken. From the results of the implementation above it was found that the accuracy of object detection using Harris Corner Detector's angle detection method is 88%. Therefore, this application can help detecting objects based on the number of corner and location detected using a Smartphone.