A sea–sky–line detection method for long wave infrared image based on improved Swin Transformer
Long wave infrared (LWIR) imaging technology is booming due to its all-weather capability. Sea–sky–line (SSL) detection based on LWIR images is a promising research in marine environment perception. However, LWIR images have been suffering the lack of rich features, challenge arises from the accurat...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/175845 |
| _version_ | 1811695045632327680 |
|---|---|
| author | Li, Chenming Cai, Chengtao Zhou, Wentao Wu, Kejun |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Li, Chenming Cai, Chengtao Zhou, Wentao Wu, Kejun |
| author_sort | Li, Chenming |
| collection | NTU |
| description | Long wave infrared (LWIR) imaging technology is booming due to its all-weather capability. Sea–sky–line (SSL) detection based on LWIR images is a promising research in marine environment perception. However, LWIR images have been suffering the lack of rich features, challenge arises from the accurate SSL detection in complex sea–sky background. In this paper, we propose a novel SSL detection method for LWIR images, which consists of three algorithms. First, a three-channel reconstruction algorithm for local images is proposed to increase the amount of SSL features. Second, an improved Swin Transformer network is presented for local image SSL identification, which improves the identification speed while ensuring accuracy. Third, a local SSL extraction algorithm is designed and applied to global SSL detection. Experimental results demonstrate that the proposed SSL detection method is more robust to complex background environments than the existing methods. As high as 98.9% average accuracy of SSL detection in LWIR images can be achieved, which outperforms all comparison methods. The extracted SSL is visually closer to the nature SSL, where the radian effect caused by camera distortion can be well fitted. Moreover, ablation studies are also conducted to validate the effectiveness of the proposed three algorithms in our method. |
| first_indexed | 2024-10-01T07:17:14Z |
| format | Journal Article |
| id | ntu-10356/175845 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T07:17:14Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | ntu-10356/1758452024-05-08T01:59:40Z A sea–sky–line detection method for long wave infrared image based on improved Swin Transformer Li, Chenming Cai, Chengtao Zhou, Wentao Wu, Kejun School of Electrical and Electronic Engineering Engineering Swin Transformer Sea–sky–line Long wave infrared (LWIR) imaging technology is booming due to its all-weather capability. Sea–sky–line (SSL) detection based on LWIR images is a promising research in marine environment perception. However, LWIR images have been suffering the lack of rich features, challenge arises from the accurate SSL detection in complex sea–sky background. In this paper, we propose a novel SSL detection method for LWIR images, which consists of three algorithms. First, a three-channel reconstruction algorithm for local images is proposed to increase the amount of SSL features. Second, an improved Swin Transformer network is presented for local image SSL identification, which improves the identification speed while ensuring accuracy. Third, a local SSL extraction algorithm is designed and applied to global SSL detection. Experimental results demonstrate that the proposed SSL detection method is more robust to complex background environments than the existing methods. As high as 98.9% average accuracy of SSL detection in LWIR images can be achieved, which outperforms all comparison methods. The extracted SSL is visually closer to the nature SSL, where the radian effect caused by camera distortion can be well fitted. Moreover, ablation studies are also conducted to validate the effectiveness of the proposed three algorithms in our method. This work was supported by the National Natural Science Foundation of China under Grant 52171332. 2024-05-08T01:59:40Z 2024-05-08T01:59:40Z 2024 Journal Article Li, C., Cai, C., Zhou, W. & Wu, K. (2024). A sea–sky–line detection method for long wave infrared image based on improved Swin Transformer. Infrared Physics and Technology, 138, 105125-. https://dx.doi.org/10.1016/j.infrared.2024.105125 1350-4495 https://hdl.handle.net/10356/175845 10.1016/j.infrared.2024.105125 2-s2.0-85184843171 138 105125 en Infrared Physics and Technology © 2024 Elsevier B.V. All rights reserved. |
| spellingShingle | Engineering Swin Transformer Sea–sky–line Li, Chenming Cai, Chengtao Zhou, Wentao Wu, Kejun A sea–sky–line detection method for long wave infrared image based on improved Swin Transformer |
| title | A sea–sky–line detection method for long wave infrared image based on improved Swin Transformer |
| title_full | A sea–sky–line detection method for long wave infrared image based on improved Swin Transformer |
| title_fullStr | A sea–sky–line detection method for long wave infrared image based on improved Swin Transformer |
| title_full_unstemmed | A sea–sky–line detection method for long wave infrared image based on improved Swin Transformer |
| title_short | A sea–sky–line detection method for long wave infrared image based on improved Swin Transformer |
| title_sort | sea sky line detection method for long wave infrared image based on improved swin transformer |
| topic | Engineering Swin Transformer Sea–sky–line |
| url | https://hdl.handle.net/10356/175845 |
| work_keys_str_mv | AT lichenming aseaskylinedetectionmethodforlongwaveinfraredimagebasedonimprovedswintransformer AT caichengtao aseaskylinedetectionmethodforlongwaveinfraredimagebasedonimprovedswintransformer AT zhouwentao aseaskylinedetectionmethodforlongwaveinfraredimagebasedonimprovedswintransformer AT wukejun aseaskylinedetectionmethodforlongwaveinfraredimagebasedonimprovedswintransformer AT lichenming seaskylinedetectionmethodforlongwaveinfraredimagebasedonimprovedswintransformer AT caichengtao seaskylinedetectionmethodforlongwaveinfraredimagebasedonimprovedswintransformer AT zhouwentao seaskylinedetectionmethodforlongwaveinfraredimagebasedonimprovedswintransformer AT wukejun seaskylinedetectionmethodforlongwaveinfraredimagebasedonimprovedswintransformer |