Euclid: Identification of asteroid streaks in simulated images using deep learning

Open Access

Table 3

Detailed detection completeness of the inter-exposure streak linking.

GT multi-streaks with 4 parts	411	23.7% of GTs
of which DL found	248	60.3%
with 4 hits	207	50.4%
with 3 hits	27	6.6%
with 2 hits	14	3.4%
GT multi-streaks with 3 parts	680	39.2% of GTs
of which DL found	416	61.2%
with 3 hits	356	52.4%
with 2 hits	60	8.8%
GT multi-streaks with 2 parts	644	37.1% of GTs
of which DL found	357	55.4%
with 2 hits	357	55.4%

Notes. The numbers are calculated for clean hits. “GT” refers to ground truth and “DL” refers to deep learning, i.e., CNN+RNN+XGBoost.

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.