Computer Vision
Robot Vision for Search-and-Rescue Facial Recognition
Facial Recognition of Human Objects Using Vision Sensors of a Robot in a Search & Rescue Operation
Helping a rescue robot find people: detecting and recognising faces from its cameras in disaster zones.
The problem
Sending responders into hazardous areas before survivors are located puts more lives at risk.
The approach
A MATLAB pipeline for face localisation and preprocessing, comparing feature-based and neural approaches and evaluating cascade detectors on standard face datasets.
The outcome
A working robot-vision pipeline that showed how onboard recognition could improve situational awareness. Final-year engineering thesis.
In detail
Final-year Mechatronics and Robotics project focused on human detection and facial recognition from robot vision sensors in post-disaster environments. The work reviewed feature-based and neural approaches, implemented MATLAB-based face localisation and preprocessing, and evaluated cascade detectors on standard face datasets. The broader objective was to improve situational awareness while reducing the need to send responders into hazardous areas before survivor locations are understood.
Built with
- Computer Vision
- Robotics
- Search & Rescue
More in Computer Vision
- Pneumonia Detection with Vision Transformers
- Few-Shot Medical Diagnosis with Model-Agnostic Meta-Learning
Next project: Aquacell PPE Research & Prototype Development
By Kanishka Ranaweera, PhD, Machine Learning Engineer in Melbourne, Victoria, Australia. hello@kanishkabuilt.me