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

Supervised deep learning was recently introduced in high-contrast imaging(HCI) through the SODINN algorithm, a convolutional neural network designed forexoplanet detection in angular differential imaging (ADI) datasets. Thebenchmarking of HCI algorithms within the Exoplanet Imaging Data Challenge(EIDC) showed that (i) SODINN can produce a high number of false positives inthe final detection maps, and (ii) algorithms processing images in a more localmanner perform better. This work aims to improve the SODINN detectionperformance by introducing new local processing approaches and adapting itslearning process accordingly. We propose NA-SODINN, a new deep learning binaryclassifier based on a convolutional neural network (CNN) that better capturesimage noise correlations in ADI-processed frames by identifying noise regimes.Our new approach was tested against its predecessor, as well as twoSODINN-based hybrid models and a more standard annular-PCA approach, throughlocal receiving operating characteristics (ROC) analysis of ADI sequences fromthe VLT/SPHERE and Keck/NIRC-2 instruments. Results show that NA-SODINNenhances SODINN in both sensitivity and specificity, especially in thespeckle-dominated noise regime. NA-SODINN is also benchmarked against thecomplete set of submitted detection algorithms in EIDC, in which we show thatits final detection score matches or outperforms the most powerful detectionalgorithms.Throughout the supervised machine learning case, this studyillustrates and reinforces the importance of adapting the task of detection tothe local content of processed images.

NA-SODINN: a deep learning algorithm for exoplanet image detection based on residual noise regimes