A new multiperspective framework for standardization and benchmarking of image dehazing algorithms
A standardization and benchmarking framework for image dehazing algorithms based on multiple perspectives is not yet available. Hence, this study proposed a new multi-perspective standardization and benchmarking framework for image dehazing algorithms. Experiments were conducted in three main phases...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 语言: | English English English |
| 出版: |
2021
|
| 主题: | |
| 在线阅读: | http://eprints.uthm.edu.my/1789/2/KARRAR%20HAMEED%20ABDUL%20KAREEM%20-%20declaration.pdf http://eprints.uthm.edu.my/1789/1/KARRAR%20HAMEED%20ABDUL%20KAREEM%20-%2024p.pdf http://eprints.uthm.edu.my/1789/3/KARRAR%20HAMEED%20ABDUL%20KAREEM%20-%20fulltext.pdf |
| 标签: |
添加标签
没有标签, 成为第一个标记此记录!
|
| id |
my-uthm-ep.1789 |
|---|---|
| record_format |
uketd_dc |
| spelling |
my-uthm-ep.17892021-10-11T08:28:31Z A new multiperspective framework for standardization and benchmarking of image dehazing algorithms 2021-05 Abdulkareem, Karrar Hameed QA71-90 Instruments and machines A standardization and benchmarking framework for image dehazing algorithms based on multiple perspectives is not yet available. Hence, this study proposed a new multi-perspective standardization and benchmarking framework for image dehazing algorithms. Experiments were conducted in three main phases. First, the image dehazing criteria were standardized based on Fuzzy-Delphi Method (FDM). Furthermore, an objective experiment was conducted to test and evaluate the selected criteria from FDM within constraints of Pearson Linear Correlation Coefficient (PLCC) and Spearman Rank Order Correlation Coefficient (SRCC). Second, an evaluation experiment was conducted to obtain a new multi-perspective decision matrix. Third, Best Worst Method (BWM) and Vlse Kriterijumska Optimizacija Kompromisno Resenje (VIKOR) methods were hybridized to determine the weight of the standardized criteria and rank the algorithms. To objectively validate the selection results, mean was applied for this purpose. To evaluate the proposed framework, two main approaches were applied. On the one hand, a standard dataset was tested on the selected criteria and image dehazing algorithms to select the best algorithm. On the other hand, a benchmarking checklist scenario was adopted to measure the feasibility of the proposed work compared to other methods. The results revealed that 11 criteria were selected as the best according to FDM stipulations. Furthermore, seven criteria had been satisfied with the PLCC and SRCC tests. Hybridization of BWM and VIKOR methods can effectively solve the challenges in the selection of the optimal algorithm. The ranking results identified Contrast Limited Adaptive Histogram Equalization (CLAHE) as the best image dehazing algorithm. Apart from that, the benchmarking checklist scenario showed the proposed framework was more effective than the benchmark study. 2021-05 Thesis http://eprints.uthm.edu.my/1789/ http://eprints.uthm.edu.my/1789/2/KARRAR%20HAMEED%20ABDUL%20KAREEM%20-%20declaration.pdf text en staffonly http://eprints.uthm.edu.my/1789/1/KARRAR%20HAMEED%20ABDUL%20KAREEM%20-%2024p.pdf text en public http://eprints.uthm.edu.my/1789/3/KARRAR%20HAMEED%20ABDUL%20KAREEM%20-%20fulltext.pdf text en validuser phd doctoral Universiti Tun Hussein Onn Malaysia Faculty of Computer Science and Information Technology |
| institution |
Universiti Tun Hussein Onn Malaysia |
| collection |
UTHM Institutional Repository |
| language |
English English English |
| topic |
QA71-90 Instruments and machines |
| spellingShingle |
QA71-90 Instruments and machines Abdulkareem, Karrar Hameed A new multiperspective framework for standardization and benchmarking of image dehazing algorithms |
| description |
A standardization and benchmarking framework for image dehazing algorithms based on multiple perspectives is not yet available. Hence, this study proposed a new multi-perspective standardization and benchmarking framework for image dehazing algorithms. Experiments were conducted in three main phases. First, the image dehazing criteria were standardized based on Fuzzy-Delphi Method (FDM). Furthermore, an objective experiment was conducted to test and evaluate the selected criteria from FDM within constraints of Pearson Linear Correlation Coefficient (PLCC) and Spearman Rank Order Correlation Coefficient (SRCC). Second, an evaluation experiment was conducted to obtain a new multi-perspective decision matrix. Third, Best Worst Method (BWM) and Vlse Kriterijumska Optimizacija Kompromisno Resenje (VIKOR) methods were hybridized to determine the weight of the standardized criteria and rank the algorithms. To objectively validate the selection results, mean was applied for this purpose. To evaluate the proposed framework, two main approaches were applied. On the one hand, a standard dataset was tested on the selected criteria and image dehazing algorithms to select the best algorithm. On the other hand, a benchmarking checklist scenario was adopted to measure the feasibility of the proposed work compared to other methods. The results revealed that 11 criteria were selected as the best according to FDM stipulations. Furthermore, seven criteria had been satisfied with the PLCC and SRCC tests. Hybridization of BWM and VIKOR methods can effectively solve the challenges in the selection of the optimal algorithm. The ranking results identified Contrast Limited Adaptive Histogram Equalization (CLAHE) as the best image dehazing algorithm. Apart from that, the benchmarking checklist scenario showed the proposed framework was more effective than the benchmark study. |
| format |
Thesis |
| qualification_name |
Doctor of Philosophy (PhD.) |
| qualification_level |
Doctorate |
| author |
Abdulkareem, Karrar Hameed |
| author_facet |
Abdulkareem, Karrar Hameed |
| author_sort |
Abdulkareem, Karrar Hameed |
| title |
A new multiperspective framework for standardization and benchmarking of image dehazing algorithms |
| title_short |
A new multiperspective framework for standardization and benchmarking of image dehazing algorithms |
| title_full |
A new multiperspective framework for standardization and benchmarking of image dehazing algorithms |
| title_fullStr |
A new multiperspective framework for standardization and benchmarking of image dehazing algorithms |
| title_full_unstemmed |
A new multiperspective framework for standardization and benchmarking of image dehazing algorithms |
| title_sort |
new multiperspective framework for standardization and benchmarking of image dehazing algorithms |
| granting_institution |
Universiti Tun Hussein Onn Malaysia |
| granting_department |
Faculty of Computer Science and Information Technology |
| publishDate |
2021 |
| url |
http://eprints.uthm.edu.my/1789/2/KARRAR%20HAMEED%20ABDUL%20KAREEM%20-%20declaration.pdf http://eprints.uthm.edu.my/1789/1/KARRAR%20HAMEED%20ABDUL%20KAREEM%20-%2024p.pdf http://eprints.uthm.edu.my/1789/3/KARRAR%20HAMEED%20ABDUL%20KAREEM%20-%20fulltext.pdf |
| _version_ |
1747830866995838976 |