Remote Sensing

Remote Sensing applies the latest Artificial Intelligence (AI) techniques to analyse and interpret Earth Observation (EO) data to gain new insights that support various resource and environmental-management sectors.

Led by Professor Kai Qin and working closely with organisations including Planet Labs and Geoplex, the Remote Sensing program is using EO data and applying the latest AI techniques to gain valuable insights on the ground.

Data capture

EO data is captured by sensors mounted on aircrafts, drones, balloons, satellites and more. There is a wide variety of data types such as optical, Light Detection and Ranging (LiDAR) or Synthetic-Aperture Radar (SAR) images. These remotely-sensed data offer complementary information and can be synergised, via data fusion techniques, to yield deeper insights of what is happening on the Earth.

Applications of Remote Sensing

Using remote sensing technology and leveraging various aspects of astrophysics, AI and computer vision have been valuable, for example, in supporting the resource sector across their mining lifecycle, from early exploration to ongoing monitoring, and environmental compliance.

Our research areas

This program has a range of research capabilities including:

• remote sensing image super-solution
• remote sensing image segmentation and classification, e.g. for multi-spectral, hyper-spectral, multi-band and multi-polarisation images
• change detection from homogenous or heterogenous remote sensing images
• EO data fusion
• research and applications of machine learning (particularly deep learning) techniques for EO data analysis
• intelligent optimisation based on swarm intelligence and evolutionary computation for remote sensing.