ORIGIN: blind detection of faint line emitters in MUSE datacubes¶
This documentation is for the version of ORIGIN currently under development.
ORIGIN is a software to perform blind detection of faint emitters in MUSE datacubes.
The algorithm is tuned to efficiently detects faint spatial-spectral emission signatures, while allowing for a stable false detection rate over the data cube and providing in the same time an automated and reliable estimation of the purity.
The algorithm implements :
1. A nuisance removal part based on a continuum subtraction combining a Discrete Cosine Transform and an iterative Principal Component Analysis,
2. A detection part based on the local maxima of Generalized Likelihood Ratio test statistics obtained for a set of spatial-spectral profiles of emission line emitters,
3. A purity estimation part, where the proportion of true emission lines is estimated from the data itself: the distribution of the local maxima in the noise only configuration is estimated from that of the local minima.
This software was initially developed by Carole Clastres, under the supervision of David Mary (Lagrange institute, University of Nice) and Roland Bacon (CRAL). It was then ported to Python by Laure Piqueras (CRAL). From November 2016 to November 2017 the software was developed by Antony Schutz (CRAL/Lagrange) and Laure. Then it was developed by Simon Conseil (CRAL), in parallel with an Octave version by David, and with contributions from Yannick Roehlly (CRAL). A lot of testing has been done also by Roland Bacon (CRAL), which also produced simulated cubes.
The project was funded by the ERC MUSICOS (Roland Bacon, CRAL).
- Example notebook
- Create the ORIGIN object
- Step01 - Preprocessing
- Step02: Define areas for PCA
- Step03: PCA threshold
- Step04: Compute PCA
- Saving and reloading the session
- Step05: Compute Correlations
- Step06: Compute detection threshold
- Step07: Detection
- Step08: Compute Spectra
- Step09: Clean results
- Step10: Create masks
- Step11: Create sources
- Final catalogs