Рақамли технологияларнинг назарий ва амалий масалалари 8-том 4-нөмір (2025) · 115-122-беттер
Optimization of Micro-Object Identification using Distributors for Correction of Distorted Image Points
Жуманов, И.И., Сафаров, Р.А., Джуманов, О.И.
Аңдатпа
A wide range of discrete models of micro-object identification and mechanisms for tracking, detecting and correcting distorted points in images have been studied. Under the conditions of internal and external influences, the presence of noise (interference), lubrication of points, a priori limitation and the presence of uncertainty. A discreptor identification mechanism has been developed based on embedding a point search graph and image representation in the form of numerical, spectral characteristics, decomposition of the Laplace matrix. The equations are solved, the eigenvalues and vectors of the Laplace matrix are determined. Parabolic, orthogonal algebraic polynomials, non-linear autoregressions, Daubechies 4 interpolation polynomial, cubic interpolation spline function, three-layer neural network are implemented. A software package for identification, recognition and classification has been implemented, which has been tested on the example of pollen grain images.
идентификацияобработка изображениймикрообъектыдескриптороптимизациякоррекция искаженных точекпрограммный комплексidentificationimage processingmicro-objects
Метадеректер дереккөзі: журналдың OAI-PMH архиві · Sindex толық мәтінді сақтамайды, дереккөзге сілтеме береді.