Study on no-reference quality assessment method of neutron radiographic images based on residual network
BackgroundThe quality of neutron radiographic images is mainly evaluated by human visual system (HVS), but HVS cannot be used as a real-time auxiliary for optimization parameters of neutron imaging systems.PurposeThis study aims to evaluate the quality of neutron radiographic images by no-reference...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Science Press
2021-07-01
|
| Series: | He jishu |
| Subjects: | |
| Online Access: | http://www.hjs.sinap.ac.cn/thesisDetails#10.11889/j.0253-3219.2021.hjs.44.070503&lang=zh |
| _version_ | 1811169661358702592 |
|---|---|
| author | QIAO Shuang LI Junhui ZHAO Chenyi ZHANG Tian |
| author_facet | QIAO Shuang LI Junhui ZHAO Chenyi ZHANG Tian |
| author_sort | QIAO Shuang |
| collection | DOAJ |
| description | BackgroundThe quality of neutron radiographic images is mainly evaluated by human visual system (HVS), but HVS cannot be used as a real-time auxiliary for optimization parameters of neutron imaging systems.PurposeThis study aims to evaluate the quality of neutron radiographic images by no-reference image quality assessment (NR-IQA) and provide an effective approach for the parameters optimization of neutron imaging systems.MethodsFirstly, the plain natural images distorted with different distortion levels and types were labelled with quality scores to construct an experimental dataset by the gradient magnitude similarity deviation (GMSD) method. Then, the residual network (ResNet) model was employed to evaluate the quality of neutron radiographic images without reference images. Finally, the goal of extracting features and assessing the quality of neutron radiographic images was achieved by training the ResNet.ResultsThe model performs well in the test set of the experimental dataset and the quality prediction of the two group authentic neutron radiographic images.ConclusionsThe proposed quality assessment method could be used for the quality prediction of neutron radiographic images. |
| first_indexed | 2024-04-10T16:46:58Z |
| format | Article |
| id | doaj.art-d60d6c5ba29e439891bf3d3520604297 |
| institution | Directory Open Access Journal |
| issn | 0253-3219 |
| language | zho |
| last_indexed | 2024-04-10T16:46:58Z |
| publishDate | 2021-07-01 |
| publisher | Science Press |
| record_format | Article |
| series | He jishu |
| spelling | doaj.art-d60d6c5ba29e439891bf3d35206042972023-02-08T00:42:02ZzhoScience PressHe jishu0253-32192021-07-01447596610.11889/j.0253-3219.2021.hjs.44.0705030253-3219（2021）07-0059-08Study on no-reference quality assessment method of neutron radiographic images based on residual networkQIAO ShuangLI JunhuiZHAO ChenyiZHANG TianBackgroundThe quality of neutron radiographic images is mainly evaluated by human visual system (HVS), but HVS cannot be used as a real-time auxiliary for optimization parameters of neutron imaging systems.PurposeThis study aims to evaluate the quality of neutron radiographic images by no-reference image quality assessment (NR-IQA) and provide an effective approach for the parameters optimization of neutron imaging systems.MethodsFirstly, the plain natural images distorted with different distortion levels and types were labelled with quality scores to construct an experimental dataset by the gradient magnitude similarity deviation (GMSD) method. Then, the residual network (ResNet) model was employed to evaluate the quality of neutron radiographic images without reference images. Finally, the goal of extracting features and assessing the quality of neutron radiographic images was achieved by training the ResNet.ResultsThe model performs well in the test set of the experimental dataset and the quality prediction of the two group authentic neutron radiographic images.ConclusionsThe proposed quality assessment method could be used for the quality prediction of neutron radiographic images.http://www.hjs.sinap.ac.cn/thesisDetails#10.11889/j.0253-3219.2021.hjs.44.070503&lang=zhneutron radiographic imagesno-reference image quality assessmentresidual networkgradient magnitude similarity deviation |
| spellingShingle | QIAO Shuang LI Junhui ZHAO Chenyi ZHANG Tian Study on no-reference quality assessment method of neutron radiographic images based on residual network He jishu neutron radiographic images no-reference image quality assessment residual network gradient magnitude similarity deviation |
| title | Study on no-reference quality assessment method of neutron radiographic images based on residual network |
| title_full | Study on no-reference quality assessment method of neutron radiographic images based on residual network |
| title_fullStr | Study on no-reference quality assessment method of neutron radiographic images based on residual network |
| title_full_unstemmed | Study on no-reference quality assessment method of neutron radiographic images based on residual network |
| title_short | Study on no-reference quality assessment method of neutron radiographic images based on residual network |
| title_sort | study on no reference quality assessment method of neutron radiographic images based on residual network |
| topic | neutron radiographic images no-reference image quality assessment residual network gradient magnitude similarity deviation |
| url | http://www.hjs.sinap.ac.cn/thesisDetails#10.11889/j.0253-3219.2021.hjs.44.070503&lang=zh |
| work_keys_str_mv | AT qiaoshuang studyonnoreferencequalityassessmentmethodofneutronradiographicimagesbasedonresidualnetwork AT lijunhui studyonnoreferencequalityassessmentmethodofneutronradiographicimagesbasedonresidualnetwork AT zhaochenyi studyonnoreferencequalityassessmentmethodofneutronradiographicimagesbasedonresidualnetwork AT zhangtian studyonnoreferencequalityassessmentmethodofneutronradiographicimagesbasedonresidualnetwork |