Alaska Project Ideas, mentored by the researchers and collaborators of University of Alaska and supported by open-source entities and enthusiasts in Alaska.

Using U-Net Model to Detect Wildfire from Satellite Imagery

This repository showcases our work on using computer vision to detect wildfires. Explore the code, model, and results of our research on wildfire prevention.

An experimental repository to build ML models and perform efficient wildfire smoke detection.

Evaluate Wildfire Environmental Impact and Assess Burn Severity Consequences using Cloud Based Geoprocessing via Earth Engine Streamlit App.

Wildfire risk assessment using remote sensing data - Prediction of Wildfires

A Wildfire Detection System that integrates machine learning models with satellite imagery, camera feeds, and weather data to predict and detect wildfires effectively.

End-to-end machine learning pipeline for the prediction of extreme and dangerous wildfires.

A collection of modules to programmatically search for/download imagery from live cam feeds across the state of California.

Utilizing Google Earth Engine and satellite imagery to identify wildfire occurrences.

Canada Wildfire Prediction Using Deep Learning

Xtinguish is an CNN Image Classfication model which helps in detecting and preventing Wildfires

This project develops an automated forest fire surveillance system using UAV drones, combining Artificial Intelligence (AI) with the YOLOv11 model optimized by TensorRT to run on NVIDIA Jetson Nano.

end-to-end pipeline to predict next-day wildfire risk from NASA FIRMS (active fires) and Meteostat weather, train LightGBM, and visualize alerts in Streamlit.

🎥🌲🔥 Improving wildfire smoke detection models by creating virtual fine-tuning data in Unreal Engine.

Low-cost wildfire monitoring using IoT sensors, edge computer vision, and deep reinforcement learning.

This repositories leverages the YOLOv5l model by ultralytics and computer vision algorithms to localize and classify some kind of anomalies that can harm wildlife animals as well as their habitate.

Quantum-Enhanced Wildfire Prediction and Resource Optimization System

Machine Learning for Wildfire Detection - MLOps

Hierarchical Vision Transformer pipeline for early wildfire detection — 97.8% binary accuracy, ablation study across 23 model variations