CONSTRUCTING METAMORPHOSIS OF IMAGES FOR THE OBJECTS ON THE BASIS OF SOLVING EULER-POINCARE EQUATIONS

This work considers comparison of two images (original and target) that represent curves correspondingly shaped of set of points in two-dimensional space. The problem can be solved by finding the initial image of diffeomorphism that would allow to overlap matching points of deformable image with tem...

Full description

Bibliographic Details
Main Author: S. V. Leichter
Format: Article
Language:Russian
Published: Siberian State Automobile and Highway University 2017-08-01
Series:Вестник СибАДИ
Subjects:
Online Access:https://vestnik.sibadi.org/jour/article/view/57
Description
Summary:This work considers comparison of two images (original and target) that represent curves correspondingly shaped of set of points in two-dimensional space. The problem can be solved by finding the initial image of diffeomorphism that would allow to overlap matching points of deformable image with template’s points. Solution to the problem is based on the method of constructing a minimized functionality, which characterizes evolution process of the diffeomorphic transformation of the image from initial to the final one and “penalty charge” for the deviation of the image path from the required trajectory. An algorithm for solving the diffeomorphic transformation equation is developed on the basis of the gradient descent method. The considered problem of comparing two images can be used for constructing optimal metamorphosis of images, when there is no exact correspondence between the target image and the final image of the diffeomorphism. The designed algorithms can be used through a biometrical system, in images and subjects classification systems, machine vision systems, images and patterns recognition, tracking systems.
ISSN:2071-7296
2658-5626