RUS  ENG
Full version
JOURNALS // Computer Optics // Archive

Computer Optics, 2015 Volume 39, Issue 5, Pages 738–743 (Mi co39)

This article is cited in 10 papers

IMAGE PROCESSING, PATTERN RECOGNITION

A preprocessing method for correlation-extremal systems

O. V. Tsvetkova, L. V. Tananykinab

a Saint Petersburg State Electrotechnical University «LETI», St. Petersburg, Russia
b Federal State Unitary Enterprise «State Research Institute of Applied Problems», St. Petersburg, Russia

Abstract: In this paper we discuss an image preprocessing method for different shooting conditions. The method can be applied in machine vision systems using a correlation-extremal mapping method. An information-theoretic method for image preprocessing based on entropy analysis is offered. The investigation of the method has shown that, when preprocessed, same-scene images obtained under different conditions have a more stable correlation coefficient than the original images.

Keywords: machine vision, entropy analysis, correlation, image preprocessing.

Received: 18.06.2015
Revised: 04.11.2015

DOI: 10.18287/0134-2452-2015-39-5-738-743



© Steklov Math. Inst. of RAS, 2024