Onset Wildfire Detection

Dr. Stephen W. Neville, University of Victoria, sneville@ece.uvic.ca, 250-721-6017

Tanka is a Vancouver-based, high-tech start-up focused on providing methods for the early detection of global wildfires. This then enables more efficient and informed fire management, containment, and risk-reduction processes.

Tanka achieves onset wildfire detection by using data analysis and machine learning methods to process incoming satellite data streams, while applying their rich historical data sets to advance and improve their products and services.

Tanka has partnered with the Matrix Institute for Applied Data Science through Dr. Stephen W. Neville’s research team to focus on:

• Developing confidence measures for various operational and in-development wildfire detection methods,
• Developing synthetic data generation approaches for assessing detection method operational envelopes,
• Address coarse quantization issues arising within the data sets, and
• Identifying potential analysis improvements and new assessment methods relative to servicing current and future opportunities.

Common to industry data science projects, this collaboration involves industry proprietary and confidential data and information.  As such, all research will be conducted within Dr. Neville’s purpose-built and network-isolated research cluster housed within his Information Security and Privacy Research (InSPiRe) lab, which includes a 380G memory, 17T storage, and a secure data analytics server.

Core to Dr. Neville’s research is converting industry-held data sets into correct and actionable knowledge, as informed by his combined background across data analytics and machine learning, large-scale software engineering, cyber-security and privacy, and high-tech entrepreneurship. Details of which can be found at Dr. Neville’s academic website and the Entrepreneurship@UVic (E@UVic) website.