No-Reference Quality Assessment of Extended Target Adaptive Optics Images Using Deep Neural Network
This paper proposes a supervised deep neural network model for accomplishing highly efficient image quality assessment (IQA) for adaptive optics (AO) images. The AO imaging systems based on ground-based telescopes suffer from residual atmospheric turbulence, tracking error, and photoelectric noise,...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/1/1 |
| _version_ | 1797358202733461504 |
|---|---|
| author | Guoqing Gao Lingxiao Li Hao Chen Ning Jiang Shuqi Li Qing Bian Hua Bao Changhui Rao |
| author_facet | Guoqing Gao Lingxiao Li Hao Chen Ning Jiang Shuqi Li Qing Bian Hua Bao Changhui Rao |
| author_sort | Guoqing Gao |
| collection | DOAJ |
| description | This paper proposes a supervised deep neural network model for accomplishing highly efficient image quality assessment (IQA) for adaptive optics (AO) images. The AO imaging systems based on ground-based telescopes suffer from residual atmospheric turbulence, tracking error, and photoelectric noise, which can lead to varying degrees of image degradation, making image processing challenging. Currently, assessing the quality and selecting frames of AO images depend on either traditional IQA methods or manual evaluation by experienced researchers, neither of which is entirely reliable. The proposed network is trained by leveraging the similarity between the point spread function (PSF) of the degraded image and the Airy spot as its supervised training instead of relying on the features of the degraded image itself as a quality label. This approach is reflective of the relationship between the degradation factors of the AO imaging process and the image quality and does not require the analysis of the image’s specific feature or degradation model. The simulation test data show a Spearman’s rank correlation coefficient (SRCC) of 0.97, and our method was also validated using actual acquired AO images. The experimental results indicate that our method is more accurate in evaluating AO image quality compared to traditional IQA methods. |
| first_indexed | 2024-03-08T14:58:29Z |
| format | Article |
| id | doaj.art-85fb26f9f56c4998831f45e990c41461 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-08T14:58:29Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-85fb26f9f56c4998831f45e990c414612024-01-10T15:08:03ZengMDPI AGSensors1424-82202023-12-01241110.3390/s24010001No-Reference Quality Assessment of Extended Target Adaptive Optics Images Using Deep Neural NetworkGuoqing Gao0Lingxiao Li1Hao Chen2Ning Jiang3Shuqi Li4Qing Bian5Hua Bao6Changhui Rao7Key Laboratory of Adaptive Optics, Chinese Academy of Sciences, Chengdu 610209, ChinaKey Laboratory of Adaptive Optics, Chinese Academy of Sciences, Chengdu 610209, ChinaKey Laboratory of Adaptive Optics, Chinese Academy of Sciences, Chengdu 610209, ChinaKey Laboratory of Adaptive Optics, Chinese Academy of Sciences, Chengdu 610209, ChinaKey Laboratory of Adaptive Optics, Chinese Academy of Sciences, Chengdu 610209, ChinaKey Laboratory of Adaptive Optics, Chinese Academy of Sciences, Chengdu 610209, ChinaKey Laboratory of Adaptive Optics, Chinese Academy of Sciences, Chengdu 610209, ChinaKey Laboratory of Adaptive Optics, Chinese Academy of Sciences, Chengdu 610209, ChinaThis paper proposes a supervised deep neural network model for accomplishing highly efficient image quality assessment (IQA) for adaptive optics (AO) images. The AO imaging systems based on ground-based telescopes suffer from residual atmospheric turbulence, tracking error, and photoelectric noise, which can lead to varying degrees of image degradation, making image processing challenging. Currently, assessing the quality and selecting frames of AO images depend on either traditional IQA methods or manual evaluation by experienced researchers, neither of which is entirely reliable. The proposed network is trained by leveraging the similarity between the point spread function (PSF) of the degraded image and the Airy spot as its supervised training instead of relying on the features of the degraded image itself as a quality label. This approach is reflective of the relationship between the degradation factors of the AO imaging process and the image quality and does not require the analysis of the image’s specific feature or degradation model. The simulation test data show a Spearman’s rank correlation coefficient (SRCC) of 0.97, and our method was also validated using actual acquired AO images. The experimental results indicate that our method is more accurate in evaluating AO image quality compared to traditional IQA methods.https://www.mdpi.com/1424-8220/24/1/1image quality assessmentAO imagesdeep neural networkpoint spread function |
| spellingShingle | Guoqing Gao Lingxiao Li Hao Chen Ning Jiang Shuqi Li Qing Bian Hua Bao Changhui Rao No-Reference Quality Assessment of Extended Target Adaptive Optics Images Using Deep Neural Network Sensors image quality assessment AO images deep neural network point spread function |
| title | No-Reference Quality Assessment of Extended Target Adaptive Optics Images Using Deep Neural Network |
| title_full | No-Reference Quality Assessment of Extended Target Adaptive Optics Images Using Deep Neural Network |
| title_fullStr | No-Reference Quality Assessment of Extended Target Adaptive Optics Images Using Deep Neural Network |
| title_full_unstemmed | No-Reference Quality Assessment of Extended Target Adaptive Optics Images Using Deep Neural Network |
| title_short | No-Reference Quality Assessment of Extended Target Adaptive Optics Images Using Deep Neural Network |
| title_sort | no reference quality assessment of extended target adaptive optics images using deep neural network |
| topic | image quality assessment AO images deep neural network point spread function |
| url | https://www.mdpi.com/1424-8220/24/1/1 |
| work_keys_str_mv | AT guoqinggao noreferencequalityassessmentofextendedtargetadaptiveopticsimagesusingdeepneuralnetwork AT lingxiaoli noreferencequalityassessmentofextendedtargetadaptiveopticsimagesusingdeepneuralnetwork AT haochen noreferencequalityassessmentofextendedtargetadaptiveopticsimagesusingdeepneuralnetwork AT ningjiang noreferencequalityassessmentofextendedtargetadaptiveopticsimagesusingdeepneuralnetwork AT shuqili noreferencequalityassessmentofextendedtargetadaptiveopticsimagesusingdeepneuralnetwork AT qingbian noreferencequalityassessmentofextendedtargetadaptiveopticsimagesusingdeepneuralnetwork AT huabao noreferencequalityassessmentofextendedtargetadaptiveopticsimagesusingdeepneuralnetwork AT changhuirao noreferencequalityassessmentofextendedtargetadaptiveopticsimagesusingdeepneuralnetwork |