Scholarly record
BOUNDARY ESTIMATES OF AN SIGNAL CLASSIFICATION EFFICIENCY
Abstract
The key objective of modern vision systems is the problem of a signal class determining since the signal features are completely determined by the type of underlying surface relative to which the observation is made. In spite of the fact that at present there are a huge number of analytic methods for the solving the indicated problem, in the vast majority of tasks based on natural images, capacities of solving it are characterized by the probabilities of erroneous and proper decisions. The exact analytic probability estimate can be calculated in particular cases only. Under more common conditions, it is advisable to use Kailath boundary ratio for a normalization of probability scores obtaining for system processing characteristics. The Kailath boundary ratio is considered both for the case of two hypotheses, and for the case of adjective hypotheses. The obtained results are illustrated for some specific examples, which deal with recognition tasks with application of radar images.
Publication Impact Profile
Publication details
References5
Dorosinskiy L.G., Synthesis and analysis of detection algorithm of distributed sources against background with distributed clutter, International Journal of Pure and Applied Mathematics, V.108, №3, Bulgaria, 2016, pp 651-658;
Fukunaga K., Introduction to statistical pattern recognition. 2nd Edition, USA, 2013, 592 p.
Sahoo P.K., Soltani S., Wong A.K.S., Cher Y.C., A surweg thresholding techniques, Computer Vision,Graphics and Image Processing, V.41, №2, USA, 1988, pp 233-260.
Yang, D., Yang, X., Liao, G., Zhu, S., Strong Clutter Suppression via RPCA in Multichannel SAR/GMTI System., IEEE Geoscience and Remote Sensing Letters, V.12, №12, India, 2015, pp 2237-2241.
Skolnik M.I., Radar handbook. Second edition., USA, 1990, 2535 p.
Citing literature
Number of times cited according to Crossref: 1
View or Download full articleAccess options
SWS access login
Login as SWS Scientific CommitteeLogin as SWS Scientific PartnerLogin as SWS AuthorAuthors and approved SWS contributors will read and export their own linked papers after identity matching by SWS profile, email and SGEM GlobalID.
For librarian assistance: [email protected]
Purchase Instant Access
- Article can be downloaded after successful payment.
- Article may be used according to SWS library access terms.
- Article cannot be redistributed.
