Skip to content

m-faezi/multi-spectral-sim

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Multi-Spectral-Sim

DOI

A Python-based simulator for generating synthetic multi-band astronomical images to evaluate and validate multi-spectral source segmentation algorithms, particularly the MMTO (Faezi et al.).

The simulation mimics actual multi-band observations of a specific section of the sky and includes various sources (stars, galaxies, and background galaxies), each with its own unique parameters, such as surface brightness, flux per unit solid angle of the image, effective radius, and axis ratio. To each frame, different Gaussian and Poissonian noises were added.

first band second band third band MMTO-no-merge MMTO-full-merge
Top row: Three-band astronomical image simulation with different parameters of surface brightness, effective radius, axis ratio, and noise distribution. Each frame has a size of 4,000^2 pixels.
Bottom row: MMTO multi-band segmentation map output, where the 'co_sim' is set to (left panel) a value greater than one or less than zero, and (right panel) the optimal value (0.94 for this data set).
MMTO GitHub

Installation

$ cd /path/to/workspace
$ git clone git@github.com:m-faezi/Multi-Spectral-Sim.git
$ cd Multi-Spectral-Sim
$ python -m venv ./venvs/Multi-Spectral-Sim
$ source ./venvs/Multi-Spectral-Sim/bin/activate
$ pip install -r requirements.txt

We acknowledge that the simulations will not reproduce all the real data complexities (e.g., different galaxy morphologies, artifacts, PSF, satellite tracks, etc.), which might lead to biases in the analysis of the completeness limits.

This simulator draws inspiration from single-band astronomical simulations used in Venhola et al..

Bibliography

About

Multi-spectral astronomical image simulator for source segmentation algorithms validations.

Topics

Resources

Stars

Watchers

Forks

Contributors

Languages