Comprehensive assessment of image quality with the use of the information criteria of homogeneity

Author(s):  N.G. Zakharchenko, candidate of Sciences, associate Professor, Belgorod University of Cooperation, Economics and Law, Belgorod, Russia, zng@bukep.ru

E.M. Mamatov, candidate of Sciences, Belgorod State University, Belgorod, Russia, mamatov@bsu.edu.ru

D.A. Yatsynyuk, Belgorod State University, Belgorod, Russia, dashayatsynyuk@gmail.com

N.V. Kolos, candidate of Sciences, associate Professor, Belgorod University of Cooperation, Economics and Law, Belgorod, Russia, kaf-ist-zav@bukep.ru

Issue:  Volume 45, №1

Rubric:  Infocommunication technologies

Annotation:  This paper reveals existing methods for estimating image quality, based on well-known criteria for evaluation of image quality with the use of the information criterion of homogeneity. Existing methods do not take into account all possible causes of quality loss and all types of images. The main goal of the study is proposing of universal effective method for evaluation of image quality. This method is based on well-known criteria for evaluation of image quality with the use of the information criterion of homogeneity. Image quality can be estimated with following criteria: degree of change for brightness, contrast and tone saturation. By themselves criteria does not give correct results for all different types of images and causes of quality loss. This paper proposes to use composition of these criteria instead of using criteria itself. Offered method allows to unify method of evaluation of image quality. Console application was developed for experimentation with offered method with use of C++ programming language and computer vision framework OpenCV. Results of this experimentation proved high performance and effectiveness of evaluation of image quality with proposed method

Keywords:  comprehensive assessment of image quality, homogeneity, degree of change for criteria, degree of change for brightness, degree of change for contrast, degree of change for tone saturation

Full text (PDF):  Download

Downloads count:  482