Deep Learning Based Fire Recognition for Wildfire Drone Automation

Deep Learning Based Fire Recognition for Wildfire Drone Automation

By: Robin Yadav

Wildfires are one of the most devastating and harmful natural disasters that occur in Canada every year. Firefighters require the best tools and equipment to fight these wildfires and limit their damage. Drones are becoming an increasingly useful asset to firefighters for wildfire monitoring and assessment. This project leverages deep learning to create a novel video-based fire detection system to add fire recognition and automation capability to drones.

Using Neural Networks and K-means Clustering for Accurate Wildfire Environmental Conditions Detection

Using Neural Networks and K-means Clustering for Accurate Wildfire Environmental Conditions Detection

By: Gurik Mangat

This study aims to utilize artificial neural networks (ANNs) integrated into an application interface to make wildfire environmental conditions detection fast and accurate for firefighters in the field. In order to achieve this, we constructed a dataset containing a list of widely accepted environmental conditions that contribute to wildfire spread and ignition and utilized a K-means clustering algorithm on NDVI imagery to analyze fuel moisture. This was integrated into an easy-to-use desktop application

Biograde Yeast: Biomanufacturing with Antifreeze Yeast on Mars

Biograde Yeast: Biomanufacturing with Antifreeze Yeast on Mars

By: Patricia Rea

Long-term settlements in space will have to be self-sufficient, producing their own resources, while keeping in mind transport constraints. Yeast is a perfect candidate for use on these missions as it is relatively resilient, well understood, and it is small and lightweight. This research is part of the development of a strain of Saccharomyces cerevisiae (S. cerevisiae) with the goal of surviving these temperatures through the use of antifreeze proteins (AFPs).

Mars Rover Biosignature Recognition and Analysis Simulation

Mars Rover Biosignature Recognition and Analysis Simulation

By: Ryan Walker and Rachel Parson

The St. Martin crater in Gypsumville, Manitoba (Figure 1), became a terrestrial analogue site due to the various unknowns surrounding its formation and composition. Our project focus was to analyze the target selection and sample triage capabilities of a rover through comparison between data gathered by an on-site team, with equipment similar to that of the Curiosity rover, and data collected by the off-site team, through looking at the photos of the site.