Research profile M.H. (Mohammed) Faezi

M.H. (Mohammed) Faezi

Project: The Low Surface Brightness Universe: Multidimensional Faint-Object Detection

Astronomy is experiencing rapid growth in data size and complexity. Future surveys, e.g., LSST, Euclid, KIDS, DESI and SKA will increase the number of available objects and deliver wide-deep images of the sky, including galaxies that are too faint to be seen today. It is not possible to manually inspect all images produced by these surveys, making computer science, machine learning algorithms, and advanced image analysis of vital importance. Such deep-imaging surveys are ideal to apply machine learning models, as they will be wider and deeper than any survey conducted before. In this new data-rich era, astronomy and computer science can benefit greatly from each other. Their synergy will lead to tools that will allow us to use the information of the Low Surface Brightness Universe hidden in the new surveys and will allow us to start uncovering a completely new parameter space.

The principal goal of this project is improvements to MTObjects , which is a Max-Tree based method for extraction of faint extended sources, to handle multi-band optical data, and extend the tool to multidimensional datasets, to work efficiently on 3D optical data.

Keywords: Machine learning, Astronomical imaging, Object detection

Fields of expertise involved: Machine learning, Astronomy, Image processing