
A Cloud and Vision-Based Navigation System for Blind People
Software as a Service
NoCategory
AI & Machine LearningClients
Mapua University
Techstack
PythonYOLOv5Raspberry PiFirebasePyQt5OpenCV
Purpose
Help visually impaired users navigate their surroundings safely using real-time object detection, with cloud synchronization so the system's captured data and status are available beyond the device itself.
Description
A Raspberry Pi-based prototype runs YOLOv5 object detection on live camera input to identify obstacles in the user's path, with alternative models including RetinaNet evaluated across design iterations. Firebase synchronizes captured detections to the cloud, and a PyQt5 desktop interface supports monitoring and configuration. Built in Python as a thesis project at Mapua University.




