Welcome to the School of Digital Science (SDS) Ecological Informatics Research Group
We leverage emerging techniques such as Artificial Intelligence, Computer Vision and Remote sensing to build robust, adaptable and resilient pipelines to solve ecological problems.
Our research is motivated by the increasing natural and anthropogenic activities that endanger forest covers in Brunei. To help forest managers, ecologists and conservationists with real-time information on the status, extent and health of forests, we are building baseline models that can be applied in other tropical regions.
Below is our current progress
- Spatial-temporal mapping of forest vegetation cover changes along highways in Brunei using deep learning techniques and Sentinel-2 images
- FCD-AttResU-Net: An improved forest change detection in Sentinel-2 satellite images using attention residual U-Net
- Spatio-Temporal Forecasting of Wildfires in the Face of Climate Change
The above research output demonstrate our commitment to solve ecological problems using emerging techniques that can be reproduced for wider applications in similar regions.
AI-Powered Models built
Square Miles Covered
Images annotated
Project Team Members
Spatial-temporal mapping of forest vegetation cover changes along highways in Brunei using deep learning techniques and Sentinel-2 images
This study leverages cutting-edge deep neural networks ad freely available Sentinel-2 satellite images to map forest alterations along highways in Brunei, whose results can be adapted to other tropical regions.
FCD-AttResU-Net: An improved forest change detection in Sentinel-2 satellite images using attention residual U-Net
This study improves the U-Net architecture for forest cover alterations by incorporating in cutting-edge residual blocks and soft attention mechnism to improve feature extraction.
Spatio-Temporal Forecasting of Wildfires in the Face of Climate Change
A supervised learning was used to train neural networks with historical weather and historical wildfire data to predict forest fires, utilizing the power of the Machine learning models based on the Call for Code Spot Challenge for Wildfires datasets provided by IBM PAIRS Geoscope.
Adapting to Climate Change: Exploring the Efficacy of Transformer Models in Forecasting Wildfires in Australia
Our approach leverages the key features of transformers such as the attention mechanism and positional encoding to prioritize efficiency and interpretability while maintaining flexibility for capturing diverse aspects of the data.
Team
Meet the team members.
Dr. Owais Ahmed Malik
Project Lead/PI
Dr. Ong Wee Hong
Team Member
Dr. Daphne Teck Ching Lai
Team Member
Kassim Kalinaki
Research Student
Rufai Yusuf Zakari
Research StudentContact
Feel free to contact us on exciting projets that leverage deep neural networks, remote sensing images and computer vision.
Location:
Universiti Brunei Darussalam, Academy of Brunei Studies, Lebuhraya Tungku, BE1410
Email:
owais.malik@ubd.edu.bn