Multi-Resolution Filtering: a method for isolating faint, extended emission in Dragonfly data and other low resolution images.
Please read the documentation and tutorial at https://mrfiltering.readthedocs.io/en/latest/.
- Subtract compact objects from low-resolution images (such as Dragonfly) to reveal low surface brightness features.
- Characterize and subtract stellar halos in Dragonfly image.
- Download corresponding high resolution image (HSC, CFHT) of given Dragonfly image.
This example shows the tidal feature of NGC 5907, described in van Dokkum et al. (2019). The images presented there used this algorithm. Full resolution Dragonfly images and MRF results can be found here. Check this notebook for more details in how to do MRF using this Python package! 🚀
This example shows how powerful MRF is in revealing low surface brightness features. The ultra-diffuse galaxy M101-DF3 is revealed by MRF after subtracting compact objects and bright star halos according to van Dokkum et al. (2019). Check this notebook for more details.
You can also use this script to run the MRF task. Take NGC 5907 as an example:
python mrf-task.py n5907_df_g.fits ngc5907_cfht_g.fits ngc5907_cfht_r.fits ngc5907-task.yaml --galcat='gal_cat_n5907.txt' --output='n5907_g'mkdir <install dir>
cd <install dir>
git clone [email protected]:AstroJacobLi/mrf.git
cd mrf
python setup.py installIf you don't have git configured, you can also download the zip file directly from https://github.com/AstroJacobLi/mrf/archive/master.zip, then unzip it and install in the same way.
To test whether mrf is installed successfully, import mrf in Python:
import mrf, os
print(os.path.isfile(os.path.join(mrf.__path__[0], 'iraf/macosx/x_images.e')))True means you have installed mrf successfully! Bravo!
You must have galsim installed in advance. You will also need unagi to download HSC images. Python>=3 is needed, but you can try whether mrf still works under python2. Check out the dependence of mrf depends from requirements.txt.
mrf is a free software made available under the MIT License by Pieter van Dokkum (initial development) and Jiaxuan Li (implementation, maintenance, and documentation). If you use this package in your work, please cite van Dokkum et al. (2019).
Many scripts and snippets are from kungpao (written by Song Huang). Qing Liu (U Toronto) provided very useful functions for constructing wide-angle PSF, Johnny Greco kindly shared his idea of the code structure. Roberto Abraham found the first few bugs of this package and provided useful solutions. Here we appreciate their help!
This package makes use of numpy, scipy, matplotlib, Astropy, sep, SWarp, GalSim, Photutils, reproject, IRAF, STSDAS. We thank the authors of these tools for their great efforts.
Copyright 2019 Pieter van Dokkum and Jiaxuan Li.