Automatic Correction of Barrel Distorted Images Using a Cascaded Evolutionary Estimator

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216275%3A25410%2F16%3A39901185" target="_blank" >RIV/00216275:25410/16:39901185 - isvavai.cz</a>

Výsledek na webu

<a href="http://www.sciencedirect.com/science/article/pii/S0020025516303103" target="_blank" >http://www.sciencedirect.com/science/article/pii/S0020025516303103</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1016/j.ins.2016.03.055" target="_blank" >10.1016/j.ins.2016.03.055</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Automatic Correction of Barrel Distorted Images Using a Cascaded Evolutionary Estimator

Popis výsledku v původním jazyce

All optical systems are to some extent burdened by one or more aberrations. Barrel distortion of an image is also an aberration. In this paper we used an innovative method to solve the problem of the centric radial distortion of a static image which serves for biometric identification of persons using 2D contour of a human hand. The method proposed uses a cascaded arrangement of two algorithms – the classic meta-heuristic, referred to as “jDE-differential evolution” and an algorithm called Covariance Matrix Adaptation Evolution Strategy. Optimizers use methods of inverse engineering and numerical mathematics to resolve the question of how to determine the correct parameters of the algebraic polynomial equation of the nth degree, by the application of which it is possible to obtain an image free of barrel distortion from an image affected by this distortion. The proposed method provides a high-quality and time-acceptable method of optimization and the option of choosing the approximation accuracy. With the use of the coefficients obtained, it is then possible to use a method called back-mapping to permanently correct the centric radial distortion aberration in the biometric scanner. Extensive experiments presented in this paper enable a better understanding of relationships, the accuracy obtained, and options of using evolutionary optimizers in a larger sense.

Název v anglickém jazyce

Automatic Correction of Barrel Distorted Images Using a Cascaded Evolutionary Estimator

Popis výsledku anglicky

All optical systems are to some extent burdened by one or more aberrations. Barrel distortion of an image is also an aberration. In this paper we used an innovative method to solve the problem of the centric radial distortion of a static image which serves for biometric identification of persons using 2D contour of a human hand. The method proposed uses a cascaded arrangement of two algorithms – the classic meta-heuristic, referred to as “jDE-differential evolution” and an algorithm called Covariance Matrix Adaptation Evolution Strategy. Optimizers use methods of inverse engineering and numerical mathematics to resolve the question of how to determine the correct parameters of the algebraic polynomial equation of the nth degree, by the application of which it is possible to obtain an image free of barrel distortion from an image affected by this distortion. The proposed method provides a high-quality and time-acceptable method of optimization and the option of choosing the approximation accuracy. With the use of the coefficients obtained, it is then possible to use a method called back-mapping to permanently correct the centric radial distortion aberration in the biometric scanner. Extensive experiments presented in this paper enable a better understanding of relationships, the accuracy obtained, and options of using evolutionary optimizers in a larger sense.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2016

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Information Sciences

ISSN

0020-0255

e-ISSN

—

Svazek periodika

366

Číslo periodika v rámci svazku

1

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

29

Strana od-do

70-98

Kód UT WoS článku

000380068900005

EID výsledku v databázi Scopus

—