A new standardisation and selection framework for real-time image dehazing algorithms from multi-foggy scenes based on fuzzy Delphi and hybrid multi-criteria decision analysis methods

Abdulkareem, Karrar Hameed and Arbaiy, Nureize and Zaidan, A. A. and Zaidan, B. B. and Albahri, O. S. and Alsalem, M. A. and M. Salih, Mahmood (2020) A new standardisation and selection framework for real-time image dehazing algorithms from multi-foggy scenes based on fuzzy Delphi and hybrid multi-criteria decision analysis methods. Neural Computing and Applications, 33. pp. 1029-1054. ISSN 0941-0643

Text
AJ 2020 (354).pdf
Restricted to Registered users only
Download (579kB) | Request a copy

Official URL: https://doi.org/10.1007/s00521-020-05020-4

Abstract

Given the rapid development of dehazing image algorithms, selecting the optimal algorithm based on multiple criteria is crucial in determining the efficiency of an algorithm. However, a sufficient number of criteria must be considered when selecting an algorithm in multiple foggy scenes, including inhomogeneous, homogenous and dark foggy scenes. However, the selection of an optimal real-time image dehazing algorithm based on standardised criteria presents a challenge. According to previous studies, a standardisation and selection framework for real-time image dehazing algorithms based on multi-foggy scenes is not yet available. To address this gap, this study proposes a new standardisation and selection framework based on fuzzy Delphi (FDM) and hybrid multi-criteria analysis methods. Experiments are also conducted in three phases. Firstly, the image dehazing criteria are standardised based on FDM. Secondly, an evaluation experiment is conducted based on standardised criteria and nine real-time image dehazing algorithms to obtain a multi-perspective matrix. Third, entropy and VIKOR methods are hybridised to determine the weight of the standardised criteria and to rank the algorithms. Three rules are applied in the standardisation process to determine the criteria. To objectively validate the selection results, mean is applied for this purpose. The results of this work can be taken into account in designing efficient methods and metrics for image dehazing.

Item Type:	Article
Uncontrolled Keywords:	Real-time image dehazing algorithms; Standardisation; Selection; Fuzzy Delphi method; Entropy; VIKOR
Subjects:	Q Science > QA Mathematics > QA299.6-433 Analysis
Divisions:	Faculty of Computer Science and Information Technology > Department of Software Engineering
Depositing User:	Mr. Shahrul Ahmad Bakri
Date Deposited:	01 Mar 2022 03:56
Last Modified:	01 Mar 2022 03:56
URI:	http://eprints.uthm.edu.my/id/eprint/6557

Actions (login required)

View Item