Free Access

Fig. 7

image

Feature importance for an RF classifier with the selection criteria set at an 8 mas minimum separation and a contrast of up to half a magnitude. Features 0 to 3 encode e1 relative errors for each exposure, features 4 to 7 are the same for the e2 component, and features 8 to 11 are the same for the size. The error bar shows the 1σ deviations of the importance of the feature on the learned dataset.

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.