Computer Vision-Based Framework for Supporting the Mobility of the Visually Impaired

Abstract

This poster presents an intelligent framework that includes several types of sensors embedded in a wearable device to support the visually impaired (VI) community. The proposed work is based on an integration of sensor-based techniques and a computer vision-based in order to introduce an efficient and economical visual device. The 98% accuracy rate of the proposed sequence is based on a wide detection view that used two camera modules. In this framework, we use several computer vision algorithms including Oriented FAST and Rotated BRIEF (ORB), k-nearest neighbors (KNN), Random sample consensus (RANSAC), and K-mean. However, the novelty of this work is the obstacle avoidance approach that is based on the image depth and fuzzy control rules. The results of our real time experiments emphasize that the proposed collision avoidance approach is able to aid the VI users in avoiding 100% of detected objects.