Statistics – Computation
Scientific paper
Nov 1983
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=1983esasp.201...69m&link_type=abstract
In ESA Statist. Methods in Astron. p 69-72 (SEE N84-19196 09-89)
Statistics
Computation
3
Algorithms, Astronomical Catalogs, Astronomical Photography, Classifications, Digital Data, Galaxies, Image Processing, Stars, Black And White Photography, Digital Techniques, Pattern Recognition
Scientific paper
A methodology for image classification developed for faint object discrimination on photographic plates is presented. It is based on the description of textural properties characterizing different categories of objects by gray-tone spatial-dependence matrices. Such matrices are computed from quantized digital subimages centered on the objects to be classified and including a convenient region of surrounding background. Statistical analyses of the co-occurrence matrices referring to preselected training sets of images permit the definition of the classification parameters. The final assignment of each object to a given class is based on objective criteria. The overall procedure requires reasonably short computation times, so that automatic classification of very large samples of objects is feasible.
No associations
LandOfFree
A classification algorithm for star-galaxy counts 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 A classification algorithm for star-galaxy counts, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and A classification algorithm for star-galaxy counts will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1485164