Enhancing Fire Detection via Computer Vision
Click the Poster to View Full Screen, Right click to save image
Grant: NSF
Christopher Eng
College:
The Dorothy and George Hennings College of Science, Mathematics, and Technology
Major:
Computer Information Systems
Faculty Research Advisor(s):
Daehan Kwak
Abstract:
Fire is a dangerous cataclysm that can stem from multiple causes from natural events, accidents, and intentional acts. there needs to be a way to address this issue. According to the US Fire Administration, there were about 372,000 residential fires and roughly 226,000 non-residential fires according to the 2020 statistics. Although being lower than the years before the issues are still very large. Additionally, the motivation for this research on fire suppression systems stems from the personal experiences of the researcher, who witnessed a friend who has lost their home to a fire. Working on this project the aim is to continually help and find ways to reduce the amount of so that it can continue its downward trend. The increasing frequency of fires, especially during the fall and winter seasons, highlights the urgent need for a solution to address fires on both small and large scales. Drawing inspiration from FLIR TECH, an automatic suppression system that utilizes AI technologies such as computer vision and machine learning, we are inspired by the idea to streamline and modularize the system. The objective of this research is to develop a program utilizing machine learning and computer vision to detect fires and use the gathered information to alert emergency services. Additionally, improvements are being made to incorporate object recognition capabilities, enabling the system to differentiate between different objects on fire and use that information to respond accordingly. To achieve these objectives, we developed a color recognition program, scaling program, object recognition program, and functionality program.