Open Access

Table 1

Characteristic comparison between classic and ML models.

Method NDetections Blim timage Requirement FP Line info
Classic 4747 2000 ≈1 ms None ≈1%
ML 12 268 200 ≈1 s GPU ≈1%

Notes. Ndetections symbolizes the number of detections for each method, Blim the brightness limit (in counts), timage the approximate time required to process each image (inference in the case of ML), Requirement is the recommended hardware requirements and FP is the percentage of false positives. Line info tells if the method is capable of directly determining the trail’s line characteristics.

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.