1702.07675
Finding strong gravitational lenses in the Kilo Degree Survey with Convolutional Neural Networks
Petrillo et al
Search for SN lenses: CNN method applied to 255 sq deg of KiDS. The current CNN is optimized to recognize lenses with Einstein radii > 1.4 arcsec, about 2x the r-band seeing in KiDS. Construct 21k color-mag selected LRG sample, of which 3 are known lenses. From this sample, CNN retrieves 761 SL candidates and correctly classifies 2 out of 3 of the known lenses (the misclassified lens has Einstein radius well below the range on which the algorithm is trained). Down-select the most reliable 56 candidates by a joint visual inspection. A consistency check on the estimated Einstein radii of the final sample of candidates suggests that it is likely composed of ~22 reliable lenses. A result consistent with what is expected from lens-statistics simulations, when applying the color-mag and Einstein-radius cuts. A conservative estimate based on the results shows that with the proposed method it should be possible to find ~100 massive LRG-galaxy lenses at z<0.4 in KiDS when completed (maximally ~2400 lenses).
No comments:
Post a Comment