No-reference quality assessment for image-based assessment of economically important tropical woods

Image Quality Assessment (IQA) is essential for the accuracy of systems for automatic recognition of tree species for wood samples. In this study, a No-Reference IQA (NR-IQA), wood NR-IQA (WNR-IQA) metric was proposed to assess the quality of wood images. Support Vector Regression (SVR) was trained...

全面介绍

Saved in:
书目详细资料
Main Authors: Rajagopal, Heshalini, Mokhtar, Norrima, Tengku Mohmed Noor Izam, Tengku Faiz, Wan Ahmad, Wan Khairunizam
格式: Article
出版: Public Library of Science 2020
主题:
在线阅读:http://eprints.um.edu.my/25348/
https://doi.org/10.1371/journal.pone.0233320
标签: 添加标签
没有标签, 成为第一个标记此记录!
id my.um.eprints.25348
record_format eprints
spelling my.um.eprints.253482020-08-14T01:59:42Z http://eprints.um.edu.my/25348/ No-reference quality assessment for image-based assessment of economically important tropical woods Rajagopal, Heshalini Mokhtar, Norrima Tengku Mohmed Noor Izam, Tengku Faiz Wan Ahmad, Wan Khairunizam TK Electrical engineering. Electronics Nuclear engineering Image Quality Assessment (IQA) is essential for the accuracy of systems for automatic recognition of tree species for wood samples. In this study, a No-Reference IQA (NR-IQA), wood NR-IQA (WNR-IQA) metric was proposed to assess the quality of wood images. Support Vector Regression (SVR) was trained using Generalized Gaussian Distribution (GGD) and Asymmetric Generalized Gaussian Distribution (AGGD) features, which were measured for wood images. Meanwhile, the Mean Opinion Score (MOS) was obtained from the subjective evaluation. This was followed by a comparison between the proposed IQA metric, WNR-IQA, and three established NR-IQA metrics, namely Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE), deepIQA, Deep Bilinear Convolutional Neural Networks (DB-CNN), and five Full Reference-IQA (FR-IQA) metrics known as MSSIM, SSIM, FSIM, IWSSIM, and GMSD. The proposed WNR-IQA metric, BRISQUE, deepIQA, DB-CNN, and FR-IQAs were then compared with MOS values to evaluate the performance of the automatic IQA metrics. As a result, the WNR-IQA metric exhibited a higher performance compared to BRISQUE, deepIQA, DB-CNN, and FR-IQA metrics. Highest quality images may not be routinely available due to logistic factors, such as dust, poor illumination, and hot environment present in the timber industry. Moreover, motion blur could occur due to the relative motion between the camera and the wood slice. Therefore, the advantage of WNRIQA could be seen from its independency from a "perfect" reference image for the image quality evaluation. © 2020 Rajagopal et al. Public Library of Science 2020 Article PeerReviewed Rajagopal, Heshalini and Mokhtar, Norrima and Tengku Mohmed Noor Izam, Tengku Faiz and Wan Ahmad, Wan Khairunizam (2020) No-reference quality assessment for image-based assessment of economically important tropical woods. PLoS ONE, 15 (5). e0233320. ISSN 1932-6203 https://doi.org/10.1371/journal.pone.0233320 doi:10.1371/journal.pone.0233320
institution Universiti Malaya
building UM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Malaya
content_source UM Research Repository
url_provider http://eprints.um.edu.my/
topic TK Electrical engineering. Electronics Nuclear engineering
spellingShingle TK Electrical engineering. Electronics Nuclear engineering
Rajagopal, Heshalini
Mokhtar, Norrima
Tengku Mohmed Noor Izam, Tengku Faiz
Wan Ahmad, Wan Khairunizam
No-reference quality assessment for image-based assessment of economically important tropical woods
description Image Quality Assessment (IQA) is essential for the accuracy of systems for automatic recognition of tree species for wood samples. In this study, a No-Reference IQA (NR-IQA), wood NR-IQA (WNR-IQA) metric was proposed to assess the quality of wood images. Support Vector Regression (SVR) was trained using Generalized Gaussian Distribution (GGD) and Asymmetric Generalized Gaussian Distribution (AGGD) features, which were measured for wood images. Meanwhile, the Mean Opinion Score (MOS) was obtained from the subjective evaluation. This was followed by a comparison between the proposed IQA metric, WNR-IQA, and three established NR-IQA metrics, namely Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE), deepIQA, Deep Bilinear Convolutional Neural Networks (DB-CNN), and five Full Reference-IQA (FR-IQA) metrics known as MSSIM, SSIM, FSIM, IWSSIM, and GMSD. The proposed WNR-IQA metric, BRISQUE, deepIQA, DB-CNN, and FR-IQAs were then compared with MOS values to evaluate the performance of the automatic IQA metrics. As a result, the WNR-IQA metric exhibited a higher performance compared to BRISQUE, deepIQA, DB-CNN, and FR-IQA metrics. Highest quality images may not be routinely available due to logistic factors, such as dust, poor illumination, and hot environment present in the timber industry. Moreover, motion blur could occur due to the relative motion between the camera and the wood slice. Therefore, the advantage of WNRIQA could be seen from its independency from a "perfect" reference image for the image quality evaluation. © 2020 Rajagopal et al.
format Article
author Rajagopal, Heshalini
Mokhtar, Norrima
Tengku Mohmed Noor Izam, Tengku Faiz
Wan Ahmad, Wan Khairunizam
author_facet Rajagopal, Heshalini
Mokhtar, Norrima
Tengku Mohmed Noor Izam, Tengku Faiz
Wan Ahmad, Wan Khairunizam
author_sort Rajagopal, Heshalini
title No-reference quality assessment for image-based assessment of economically important tropical woods
title_short No-reference quality assessment for image-based assessment of economically important tropical woods
title_full No-reference quality assessment for image-based assessment of economically important tropical woods
title_fullStr No-reference quality assessment for image-based assessment of economically important tropical woods
title_full_unstemmed No-reference quality assessment for image-based assessment of economically important tropical woods
title_sort no-reference quality assessment for image-based assessment of economically important tropical woods
publisher Public Library of Science
publishDate 2020
url http://eprints.um.edu.my/25348/
https://doi.org/10.1371/journal.pone.0233320
_version_ 1680857021887283200
score 13.250246