Sustainable Wildfire Prevention Using RPAS and Computer Vision

This research project in Canada focuses on sustainable natural resource management, particularly in forest areas. By integrating Remotely Piloted Aircraft Systems (RPAS) and Computer Vision (CV), the project aims to improve forest fire prevention and management. Collaborating with industry partners like Spexi Geospatial, the team combines academic research with practical solutions to enhance AI-driven environmental strategies. The project develops novel approaches for forest fire detection, using orthogonal drone images labeled with a unique fire risk system. A key innovation is the use of raw (LOG) images for training CV models, considering varying natural light conditions, over conventional RGB images. The research also explores the use of oblique drone images for detailed post-wildfire assessments, offering high-resolution views of vegetation and terrain structures that are vital for thorough evaluations. This pioneering approach leverages drone-based oblique imagery for both forest fire prevention and post-event analysis.

Faculty Supervisor:

Michal Aibin

Student:

Partner:

Spexi Geospatial

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

British Columbia Institute of Technology

Program: