Open Access

Table 2

Recall and precision by flux interval for different models.

Models Precision Recall by Flux Interval (mCrab)
0-0.2 0.2-0.4 0.4-0.6 0.6-0.8 0.8-1 1-1.5 1.5-2 2-3 3-5 5-10 10+
SExtractor 0.63 0.0085 0.0334 0.1037 0.1270 0.2125 0.2829 0.3560 0.3433 0.3521 0.4262 0.5385
PSF-filter + Gauss-fit 0.04 0.0159 0.0233 0.0317 0.0123 0.0141 0.0047 0.0105 0.0115 0.0128 0.0189 0.0143
Bayesian PSF likelihood 0.15 0.0102 0.0155 0.0378 0.0646 0.0845 0.1327 0.2316 0.3218 0.4306 0.5660 0.5571
Baseline256 0.81 0.0003 0.0003 0.0014 0.0196 0.0330 0.0697 0.2007 0.2892 0.4239 0.5847 0.6404
Baseline512 0.08 0.0106 0.0281 0.0611 0.1155 0.1250 0.2510 0.3893 0.4869 0.4714 0.7191 0.6478
Baseline512 0.50 0.0034 0.0111 0.0341 0.0934 0.1753 0.1966 0.3166 0.4720 0.4640 0.7131 0.6478
Baseline 0.50 0.0041 0.0123 0.0380 0.1057 0.1753 0.2147 0.3248 0.4720 0.5115 0.7072 0.7167
Framework 0.08 0.1803 0.2612 0.3014 0.3411 0.5239 0.7808 0.8090 0.9198 0.9313 0.9015 0.9015
Framework 0.52 0.1011 0.1674 0.2818 0.3281 0.4914 0.7236 0.8795 0.9198 0.9313 0.8855 0.8855
Framework 0.90 0.0048 0.0189 0.1840 0.2179 0.4122 0.6891 0.8505 0.9051 0.9313 0.8855 0.8855

Notes. Performance evaluation of precision and recall across different flux intervals. The proposed framework consistently achieves higher recall under the same precision levels, particularly in the faint-source regime, compared with benchmark methods.

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.