Safepath: Empowering the Visually Impaired with Voice Based Navigation
The Real-Time Object Discovery and Navigation Support for the Blind is a technological advancement aimed at improving the mobility and safety of the blind. A camera operates through the YOLOv8 model of deep learning object detection and the pyttsx 3 library for live object detection. By detecting objects such as people, obstacles and vehicles and dictating their respective positions (left, right or front), the system suggests directions to the user through voice feedback. With an adjustable setting, the system provides overcrowding detection in order to grant warnings if people were detected at a range higher than a fixed threshold so that users can avoid those unsafe areas. Very much in real-time, it aligns a user with feedback coming in the same instant. An interactive Tkinter interface to the system allows a visually impaired person to feel free in exploring and safely maneuvering through their environment.
Authors
- Akella Lakshmana Sai Srikar
- B. Sharmila (ORCID: https://orcid.org/0000-0003-3382-8237)
- D J Deepak
- J. Hema Pravallika
- SAI KIRAN NEMALAPURI
Institutions
- Aditya Birla (India) (IN)
Publication Details
- Published
- 2025-06-05
- DOI
- https://doi.org/10.1109/wams64402.2025.11159045
- Primary Topic
- Tactile and Sensory Interactions
- Type
- article
- Field-Weighted Citation Impact
- 0.00