An Efficient and Real-Time Emergency Exit Detection Technology for the Visually Impaired People Based on YOLOv5
Sihao Chen, Zhi Yu, Zhongda Qi, Wei Wang · 2022 IEEE International Conference on Electrical Engineering, Big Data and Algorithms (EEBDA) · 2022
Obtaining information from the surroundings has always been a great difficulty problem for visually impaired people. One of the most important problem is when facing an emergency, their security cannot be guaranteed. Whereas, there are no efficient assistant tools to help them recognize the emergency exit, which bring them great sense of fear. To get out this predicament, we adopt mature IFLYTEK voice recognition API and self-trained YOLOv5 object detection technology to purpose a emergency exit detection mini-program based on WeChat platform which can recognize the emergency exit signs by just acquiring a picture from the mobile phone camera. With the assistance of this mini-program, the safety of the visually impaired can be greatly improved.