26 March 2016 Change detection methods for distinction task of stochastic textures based on nonparametric method
Author Affiliations +
Proceedings Volume 9807, Optical Technologies for Telecommunications 2015; 98070W (2016) https://doi.org/10.1117/12.2234419
Event: XIII International Scientific and Technical Conference on Optical Technologies in Telecommunications, 2015, Ufa, Russian Federation
Abstract
The following article describes use of nonparametric method. This method is to be used to find multivariate change of random processes for image processing. It is aimed to find the borders of irregular phenomena in front of terrain. The current task of finding change and evaluation of a change point in consequent setting proposes test statistics based on value of sampling characteristic functions. The relevant criterion in a wide range of alternatives has a predetermined assessment of the asymptotic significance level. Current work also proposes an algorithm of texture segmentation for two-dimensional case. This algorithm is given as a consequence of processing operations in columns and rows of test statistics values, obtained during scanning of images. The test results are quoted.
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Albert Kh. Sultanov, Albert Kh. Sultanov, } "Change detection methods for distinction task of stochastic textures based on nonparametric method", Proc. SPIE 9807, Optical Technologies for Telecommunications 2015, 98070W (26 March 2016); doi: 10.1117/12.2234419; https://doi.org/10.1117/12.2234419
PROCEEDINGS
5 PAGES


SHARE
Back to Top