Computer Science – Learning
Scientific paper
Dec 2005
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2005aspc..345..362m&link_type=abstract
From Clark Lake to the Long Wavelength Array: Bill Erickson's Radio Science ASP Conference Series, Vol. 345, Proceedings of the
Computer Science
Learning
2
Scientific paper
The aim of the VLA Low-Frequency Sky Survey (VLSS) is to map the entire sky north of -30° declination at a frequency of 74 MHz. The principal data products are a set of continuum images (14° × 14°) and a catalog of the discrete sources detected. The VLSS catalog is created by fitting elliptical Gaussians to all of the sources that are detected at the 5 sigma level or higher. The images and the source catalog are publicly available. We have compared user defined subsets of the fields with the public catalog. We find that by eye we can detect more sources than appear in the VLSS catalog. To improve on the detection rate we have implemented machine learning algorithms. We find that for the test fields our genetic algorithm provides roughly a 40% increase in the number of sources detected as compared to the elliptical Gaussian method.
Junor William
Lazio Joseph T. W.
McGowan Katherine E.
No associations
LandOfFree
Detecting Radio Sources with Machine Learning does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Detecting Radio Sources with Machine Learning, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Detecting Radio Sources with Machine Learning will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1034220