entropy
Article
Entropy-Based Combined Metric for Automatic Objective
Quality Assessment of Stitched Panoramic Images
Krzysztof Okarma
1,
* , Wojciech Chlewicki
1
, Mateusz Kopytek
1
, Beata Marciniak
2
and Vladimir Lukin
3
Citation: Okarma, K.; Chlewicki, W.;
Kopytek, M.; Marciniak, B.; Lukin, V.
Entropy-Based Combined Metric for
Automatic Objective Quality
Assessment of Stitched Panoramic
Images. Entropy 2021, 23, 1525.
https://doi.org/10.3390/e23111525
Academic Editors: Michal Choras,
Robert Burduk, Agata Gielczyk, Rafal
Kozik and Tomasz Marciniak
Received: 8 October 2021
Accepted: 15 November 2021
Published: 17 November 2021
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
1
Department of Signal Processing and Multimedia Engineering, West Pomeranian University of Technology in
Szczecin, 70-313 Szczecin, Poland; wojciech.chlewicki@zut.edu.pl (W.C.); km46880@zut.edu.pl (M.K.)
2
Faculty of Telecommunications, Computer Science and Electrical Engineering, Bydgoszcz University of
Science and Technology, 85-796 Bydgoszcz, Poland; beata.marciniak@pbs.edu.pl
3
Department of Information and Communication Technologies, National Aerospace University,
61070 Kharkov, Ukraine; lukin@ai.kharkov.com
* Correspondence: okarma@zut.edu.pl
Abstract: Quality assessment of stitched images is an important element of many virtual reality and
remote sensing applications where the panoramic images may be used as a background as well as for
navigation purposes. The quality of stitched images may be decreased by several factors, including
geometric distortions, ghosting, blurring, and color distortions. Nevertheless, the specificity of such
distortions is different than those typical for general-purpose image quality assessment. Therefore,
the necessity of the development of new objective image quality metrics for such type of emerging
applications becomes obvious. The method proposed in the paper is based on the combination
of features used in some recently proposed metrics with the results of the local and global image
entropy analysis. The results obtained applying the proposed combined metric have been verified
using the ISIQA database, containing 264 stitched images of 26 scenes together with the respective
subjective Mean Opinion Scores, leading to a significant increase of its correlation with subjective
evaluation results.
Keywords: image quality assessment; stitched images; panoramic images; image analysis; image entropy
1. Introduction
Panoramic images, constructed as a result of image stitching operation conducted for
a series of constituent images with partially overlapping regions, may suffer from various
distortions, including blur, ghosting artifacts, and quite well visible geometric and color
distortions. The presence of such issues decreases the perceived image quality and in some
cases may be unacceptable from an aesthetic point of view. Although modern cameras and
smartphones are usually equipped with software functions making it possible to properly
register the overlapping areas of individual photos to create panoramic images, some
additional requirements should be fulfilled by users during the recording to prevent such
problems. Nevertheless, the growing availability of software and hardware solutions causes
higher popularity of panoramic images which may be useful, e.g., as wide background
images, in virtual reality scenarios, as well as in mobile robotics for the Visual Simultaneous
Localization and Mapping (VSLAM) applications.
Considering the modern applications of image stitching and image registration algo-
rithms, related to the use of cameras mounted on mobile robots, the quality of obtained
panoramic images is very important due to potential errors in vision-based control of their
motion. In the case of decreased image quality, such images might be removed from the
analysis to prevent their influence on the robot’s control. Another interesting direction
of such research in mobile robotics concerns the fusion of images acquired by unmanned
aerial vehicles (UAVs) [1,2].
One of the most relevant factors, influencing the final quality of the panoramic images,
is the appropriate choice of distinctive image features used to match the same regions
Entropy 2021, 23, 1525. https://doi.org/10.3390/e23111525 https://www.mdpi.com/journal/entropy