English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Context. The availability of large bandwidth receivers for millimeter radiotelescopes allows the acquisition of position-position-frequency data cubesover a wide field of view and a broad frequency coverage. These cubes containmuch information on the physical, chemical, and kinematical properties of theemitting gas. However, their large size coupled with inhomogenoussignal-to-noise ratio (SNR) are major challenges for consistent analysis andinterpretation.Aims. We search for a denoising method of the low SNR regions ofthe studied data cubes that would allow to recover the low SNR emission withoutdistorting the signals with high SNR.Methods. We perform an in-depth dataanalysis of the 13 CO and C 17 O (1 -- 0) data cubes obtained as part of theORION-B large program performed at the IRAM 30m telescope. We analyse thestatistical properties of the noise and the evolution of the correlation of thesignal in a given frequency channel with that of the adjacent channels. Thisallows us to propose significant improvements of typical autoassociative neuralnetworks, often used to denoise hyperspectral Earth remote sensing data.Applying this method to the 13 CO (1 -- 0) cube, we compare the denoised datawith those derived with the multiple Gaussian fitting algorithm ROHSA,considered as the state of the art procedure for data line cubes.Results. Thenature of astronomical spectral data cubes is distinct from that of thehyperspectral data usually studied in the Earth remote sensing literaturebecause the observed intensities become statistically independent beyond ashort channel separation. This lack of redundancy in data has led us to adaptthe method, notably by taking into account the sparsity of the signal along thespectral axis. The application of the proposed algorithm leads to an increaseof the SNR in voxels with weak signal, while preserving the spectral shape ofthe data in high SNR voxels.Conclusions. The proposed algorithm that combines adetailed analysis of the noise statistics with an innovative autoencoderarchitecture is a promising path to denoise radio-astronomy line data cubes. Inthe future, exploring whether a better use of the spatial correlations of thenoise may further improve the denoising performances seems a promising avenue.In addition,

Deep learning denoising by dimension reduction: Application to the ORION-B line cubes