Fig. 1

The three stages of our supervised detection framework. Panel a: labeled data generation step. The ADI sequence and off-axis PSF template are examples of VLT/SPHERE data. Panel b: model training step for the case of SODINN. SODIRF uses a random forest classifier instead of a deep neural network. Panel c: evaluation of the trained model on the original cube and shows the schematic representation of the output detection map.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.