NoiseChisel: non-parametric detection and analysis of astronomical targets

Mohammad Akhlagi

CRAL

Astronomical instrumentation has greatly advanced over the last 40 years: with digital detectors, space telescopes and +8m class ground-based telescopes for example. However, the signal-based detection paradigm (for example from Petrosian or Kron in the 1970s, mostly used as implemented in SExtractor from the mid-1990s) is still the dominant method of low-level data analysis: detection, segmentation and measurements or catalog production. In this talk, after reviewing the major systematic biases regarding astronomical object detection that is inherent to the signal-based paradigm, I will introduce a fundamentally new noise-based detection paradigm for detecting signal that may have extremely low signal-to-noise ratios. With thresholds that are far below the Sky value, and non-parametric expansion into noise, it is successfully able to detect very diffuse and irregularly shaped signal in noise (galaxies). The software implementation is called NoiseChisel. The talk will continue with an introduction to NoiseChisel’s parent software: GNU Astronomy Utilities (or Gnuastro). It is a large package of useful programs and libraries for astronomical data analysis directly on the command-line without the need to use mini-environments like IRAF or Python. It fully conforms with the GNU Coding Standards for easy integration into all Unix-like operating systems (GNU/Linux and MacOS for example) with familiar installation, usage and documentation. The talk will conclude with a discussion on reproducible scientific papers and a solution that has already been implemented to have exactly reproducible scientific papers.