Forest fires have significant implications for the Earth’s ecological balance, causing widespread devastation and posing formidable challenges for containment once they propagate. The development of computer vision methods holds promise in facilitating the timely identification of forest fire risks, thereby preventing potential economic losses. In our study conducted in various regions in British Columbia, we utilized image data captured by unmanned aerial vehicles (UAVs) and computer vision methods to detect various types of trees, including alive trees, debris (logs on the ground), beetle- and fire-impacted trees, and dead trees that pose a risk of a forest fire. We then designed and implemented a novel sliding window technique to process large forest areas as georeferenced orthogonal maps. The model demonstrates proficiency in identifying various tree types, excelling in detecting healthy trees with precision and recall scores of 0.904 and 0.848, respectively. Its effectiveness in recognizing trees killed by beetles is somewhat limited, likely due to the smaller number of examples available in the dataset. After the tree types are detected, we generate color maps, indicating different fire risks to provide a new tool for fire managers to assess and implement prevention strategies. This study stands out for its integration of UAV technology and computer vision in forest fire risk assessment, marking a significant step forward in ecological protection and sustainable forest management.
Object detection is a fundamental part of computer vision, with a wide range of real-world applications. It involves the detection of various objects in digital images or video. In this paper, we propose a proof of concept usage of computer vision algorithms to improve the maintenance of railway tracks operated by Via Rail Canada. Via Rail operates about 500 trains running on 12,500 km of tracks. These tracks pass through long stretches of sparsely populated lands. Maintaining these tracks is challenging due to the sheer amount of resources required to identify the points of interest (POI), such as growing vegetation, missing or broken ties, and water pooling around the tracks. We aim to use the YOLO algorithm to identify these points of interest with the help of aerial footage. The solution shows promising results in detecting the POI based on unmanned aerial vehicle (UAV) images. Overall, we achieved a precision of 74% across all POI and a mean average precision @ 0.5 (mAP @ 0.5) of 70.7%. The most successful detection was the one related to missing ties, vegetation, and water pooling, with an average accuracy of 85% across all three POI.