Statistics
Scientific paper
Dec 1992
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=1992aas...181.9901z&link_type=abstract
American Astronomical Society, 181st AAS Meeting, #99.01; Bulletin of the American Astronomical Society, Vol. 24, p.1275
Statistics
Scientific paper
Source detection from imaging data is usually performed by calculating a local signal-to-noise ratio based on Gaussian statistics. In the photon-limited case this assumption is inadequate and leads to fewer source detections than if Poisson noise is assumed. We propose an alternative method that corrects this by calculating the expected histogram of counts of the background noise and then comparing it to the histogram of counts of the image. If sources are present in the image, they will appear as differences between the two histograms. Monte-Carlo simulations were used to compare the results obtained by our method with those of conventional methods. They show that our method is more suitable for faint sources (or with S/N < 5) since it produces much fewer spurious detections. We applied it to data from the EINSTEIN X-Ray Observatory (IPC detector) in search of pulsars that might be associated with new x-ray sources. A sample of 1900 candidate sources was then obtained at the Arecibo radiotelescope declination range to be checked for radio pulsations.
Cordes James M.
Wasserman Ira
Zepka A. F.
No associations
LandOfFree
An New Method for Source Detection from Photon-limited Imaging Data 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 An New Method for Source Detection from Photon-limited Imaging Data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and An New Method for Source Detection from Photon-limited Imaging Data will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1111896