NVIDIA Disaster Risk Monitoring Using Satellite Imagery
Interested in disaster risk management? Curious about how you can use Deep Learning models to monitor potential disasters? This free, self-paced, online course developed by NVIDIA Deep Learning Institute jointly with the United Nations Satellite Centre (UNOSAT) is for you!
The course teaches participants to build and deploy a deep learning model built with different frameworks which uses satellite imagery to detect natural disasters – specifically flood events. The use of deep learning models for disaster risk management are advantageous because they lower costs, increase efficiency and increase effectiveness of disaster risk monitoring.
Prerequisites
In order to take part in this course, participants must already be competent in Python 3 programming language. They are also required to have a basic understanding of Machine Learning and Deep Learning concepts and pipelines, as well as interest in manipulating satellite imagery.
Learning outcomes
By taking part in this course, participants will learn:
- Implementing a machine learning workflow for disaster management solutions
- Processing large satellite imagery data using hardware accelerated tools
- Cost-efficiently build deep learning segmentation models by applying transfer-learning
- Using deep learning models for real-time monitoring and analysis
- Detecting and responding to flood events by using deep learning-based model inference
