{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,5]],"date-time":"2026-10-05T11:52:57Z","timestamp":1791201177257,"version":"4.2.0"},"reference-count":52,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T00:00:00Z","timestamp":1783123200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,10,5]],"date-time":"2026-10-05T00:00:00Z","timestamp":1791158400000},"content-version":"vor","delay-in-days":93,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Sci Rep"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Intelligent landscape recognition sits at the heart of smart coastal tourism, yet pushing vision Transformers onto resource-constrained edge hardware remains awkward\u2014the self-attention mechanism scales quadratically with token count, and marine atmospheric haze quietly erodes recognition accuracy long before it is noticed in the laboratory. We propose a hybrid lightweight architecture that swaps standard self-attention in the early stages for depthwise separable convolution token mixers, switches to kernel-based linear attention in the deeper stages, and follows an adaptive channel reduction policy that trims feature dimensions where redundancy is empirically highest. A Fog-Aware Feature Calibration module, motivated by the Koschmieder atmospheric scattering model, is embedded between the convolutional and attention stages so that learned dehazing happens inside the network rather than as a separate preprocessing pass. Training proceeds through a three-phase pipeline that interleaves progressive structured pruning with temperature-scheduled knowledge distillation from a Swin-Base teacher; the composite objective shrinks the model to 4.8 M parameters and 0.91 GFLOPs without abrupt accuracy collapse. On a newly collected 17,565-image dataset spanning eight coastal scene categories across six shoreline regions of Zhejiang Province, our model reaches 92.6%\u2009\u00b1\u20090.4% Top-1 accuracy (mean\u2009\u00b1\u2009std over three independent runs), the best result among all baselines below 6 M parameters, including the recently released MobileViT-v2, FastViT, TinyViT, EfficientViT and RepViT. On an NVIDIA Jetson Orin Nano with TensorRT INT8 optimisation the system delivers 18.3 ms latency (54.6 FPS) within a 7.4 W average power envelope, and a 200 m visibility heavy-fog subset that mixes synthetic and real captures shows an 85.7% accuracy retention rate\u20146.8 percentage points above the strongest baseline. Cross-dataset zero-shot tests on Fujian, Hainan and the Places365 coastal subset, together with deployment benchmarks on Jetson Nano 4 GB and Coral Edge TPU, suggest the design transfers reasonably\u2014though not effortlessly\u2014beyond the original collection sites.<\/jats:p>","DOI":"10.1038\/s41598-026-59312-1","type":"journal-article","created":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T08:55:25Z","timestamp":1783155325000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A fog-aware lightweight vision transformer architecture for real-time coastal landscape recognition on edge devices"],"prefix":"10.1038","volume":"16","author":[{"given":"Yanyan","family":"He","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dekun","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yibo","family":"Hu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qiang","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,7,4]]},"reference":[{"key":"59312_CR1","unstructured":"Chen, Y., Huang, Z., Peng, C. Digital transformation and sustainable development of coastal cultural tourism: Evidence from Zhejiang province. Ocean Coastal Manage., 245, (2023)."},{"key":"59312_CR2","unstructured":"Wang, L., Zhang, Y., Li, H. Smart tourism ecosystem construction driven by digital transformation: A framework and case study from eastern China. Tourism Manage., 95, (2023)."},{"key":"59312_CR3","doi-asserted-by":"publisher","first-page":"104245","DOI":"10.1016\/j.tourman.2020.104245","volume":"83","author":"X Li","year":"2021","unstructured":"Li, X., Law, R., Xie, J. & Wang, S. Review of tourism forecasting research with internet data. Tourism Manage 83, 104245 (2021).","journal-title":"Tourism Manage"},{"issue":"2","key":"59312_CR4","first-page":"2455","volume":"80","author":"C Li","year":"2021","unstructured":"Li, C., Guo, J. & Hu, G. Visibility restoration of single image captured in dust and haze weather conditions. Multimedia Tools Appl. 80(2), 2455\u20132471 (2021).","journal-title":"Multimedia Tools Appl."},{"key":"59312_CR5","unstructured":"Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., Houlsby, N. An image is worth 16x16 words: Transformers for image recognition at scale. In Proceedings of the international conference on learning representations (ICLR), (2021)."},{"key":"59312_CR6","unstructured":"Touvron, H., Cord, M., Douze, M., Massa, F., Sablayrolles, A., J\u00e9gou, H. Training data-efficient image transformers and distillation through attention. In Proceedings of the 38th international conference on machine learning (ICML), vol. 139, pp. 10347\u201310357, (2021)."},{"key":"59312_CR7","doi-asserted-by":"crossref","unstructured":"Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., Guo, B. Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE\/CVF international conference on computer vision (ICCV), pp. 10012\u201310022, (2021).","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"59312_CR8","unstructured":"Mehta, S., Rastegari, M. MobileViT: Light-weight, general-purpose, and mobile-friendly vision Transformer. In Proceedings of the international conference on learning representations (ICLR), (2022)."},{"key":"59312_CR9","doi-asserted-by":"crossref","unstructured":"Li, Y., Yuan, G., Wen, Y., Hu, J., Evangelidis, G., Tulyakov, S., Wang, Y., Ren, J. EfficientFormer: Vision transformers at MobileNet speed. In Advances in neural information processing systems (NeurIPS), vol. 35, (2022).","DOI":"10.52202\/068431-0940"},{"key":"59312_CR10","doi-asserted-by":"crossref","unstructured":"Yu, L., Zhang, W., Li, H. X-Pruner: Explainable pruning for vision transformers. In Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition (CVPR), pp. 24355\u201324363, (2023).","DOI":"10.1109\/CVPR52729.2023.02333"},{"key":"59312_CR11","unstructured":"Hinton, G., Vinyals, O., Dean, J., Distilling the knowledge in a neural network. In NIPS Deep learning and representation learning workshop, (2015)."},{"key":"59312_CR12","unstructured":"NVIDIA Corporation, TensorRT: Programmable inference accelerator, NVIDIA Developer Documentation, (2023)."},{"issue":"4","key":"59312_CR13","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1145\/1498765.1498785","volume":"52","author":"S Williams","year":"2009","unstructured":"Williams, S., Waterman, A. & Patterson, D. Roofline: An insightful visual performance model for multicore architectures. Commun. ACM 52(4), 65\u201376 (2009).","journal-title":"Commun. ACM"},{"key":"59312_CR14","unstructured":"Tan, M., Le, Q. V., EfficientNet: Rethinking model scaling for convolutional neural networks. In Proceedings of the 36th international conference on machine learning (ICML), vol. 97, pp. 6105\u20136114, (2019)."},{"key":"59312_CR15","doi-asserted-by":"crossref","unstructured":"Sumbul, G., Charfuelan, M., Demir, B., Markl, V. BigEarthNet: A large-scale benchmark archive for remote sensing image understanding. In Proceedings of the IEEE international geoscience and remote sensing symposium (IGARSS), pp. 5901\u20135904, (2019).","DOI":"10.1109\/IGARSS.2019.8900532"},{"issue":"12","key":"59312_CR16","doi-asserted-by":"publisher","first-page":"1284","DOI":"10.1109\/TLA.2025.11231229","volume":"23","author":"RDA Raj","year":"2025","unstructured":"Raj, R. D. A. et al. AAPN-Tiny: A compact edge-deployable adaptive attention pyramid architecture for multi-class fault diagnosis in solar photovoltaic modules. IEEE Lat. Am. Trans. 23(12), 1284\u20131296 (2025).","journal-title":"IEEE Lat. Am. Trans."},{"key":"59312_CR17","doi-asserted-by":"publisher","DOI":"10.1016\/j.compeleceng.2026.110979","volume":"131","author":"BSV Vinayak","year":"2026","unstructured":"Vinayak, B. S. V. et al. Hierarchical mobile-dense convolutional architecture for tampered image detection using focal optimization with quantized edge TPU deployment. Comput. Electr. Eng. 131, 110979 (2026).","journal-title":"Comput. Electr. Eng."},{"key":"59312_CR18","doi-asserted-by":"publisher","DOI":"10.1016\/j.epsr.2025.112451","volume":"252","author":"K Himavarshini","year":"2026","unstructured":"Himavarshini, K. et al. TinyTripleNet: A lightweight architecture for solar photovoltaic fault detection optimized for edge tensor processing unit hardware deployment. Electric Power Sys. Res. 252, 112451 (2026).","journal-title":"Electric Power Sys. Res."},{"key":"59312_CR19","doi-asserted-by":"publisher","DOI":"10.1016\/j.compeleceng.2025.110645","volume":"128","author":"BS Gangarapu","year":"2025","unstructured":"Gangarapu, B. S. et al. Lightweight spatial attention pyramid network-based image forgery detection optimized for real-time edge TPU deployment. Comput. Electr. Eng. 128, 110645 (2025).","journal-title":"Comput. Electr. Eng."},{"issue":"6","key":"59312_CR20","volume":"1","author":"RMR Yanamala","year":"2025","unstructured":"Yanamala, R. M. R., Pallakonda, A. & Amar Raj, R. D. Pruned and low-rank optimized tiny residual architecture for solar photovoltaic fault classification on edge TPU. J. Energy Resour. Technol. Part A Sustain. Renew. Energy 1(6), 061305 (2025).","journal-title":"J. Energy Resour. Technol. Part A Sustain. Renew. Energy"},{"key":"59312_CR21","volume":"27","author":"A Pallakonda","year":"2025","unstructured":"Pallakonda, A. et al. Lightweight hierarchical spatial feature extraction and sequential modeling for PV fault detection using pyramid network and GRU for edge applications. Energy Convers. Manage. X 27, 101293 (2025).","journal-title":"Energy Convers. Manage. X"},{"issue":"5","key":"59312_CR22","first-page":"3894","volume":"10","author":"Z Chen","year":"2023","unstructured":"Chen, Z., Xu, L. & Li, Q. Edge intelligence for smart tourism: Opportunities and challenges. IEEE Internet Things J. 10(5), 3894\u20133910 (2023).","journal-title":"IEEE Internet Things J."},{"issue":"1","key":"59312_CR23","doi-asserted-by":"publisher","first-page":"492","DOI":"10.1109\/TIP.2018.2867951","volume":"28","author":"B Li","year":"2019","unstructured":"Li, B. et al. Benchmarking single-image dehazing and beyond. IEEE Trans. Image Process. 28(1), 492\u2013505 (2019).","journal-title":"IEEE Trans. Image Process."},{"key":"59312_CR24","doi-asserted-by":"crossref","unstructured":"Guo, S. The construction of smart tourism city and digital marketing of cultural tourism industry under network propaganda strategy. Secur. Commun. Netw., 2022, (2022).","DOI":"10.1155\/2022\/4932415"},{"key":"59312_CR25","doi-asserted-by":"crossref","unstructured":"Devlin, J., Chang, M.-W., Lee, K., Toutanova, K. BERT: Pre-training of deep bidirectional Transformers for language understanding. In Proceedings of the conference of the North American chapter of the association for computational linguistics (NAACL), pp. 4171\u20134186, (2019).","DOI":"10.18653\/v1\/N19-1423"},{"key":"59312_CR26","unstructured":"Ba, J. L., Kiros, J. R., Hinton, G. E., Layer normalization. In Advances in neural information processing systems (NeurIPS) deep learning symposium, 2016."},{"key":"59312_CR27","first-page":"5998","volume":"30","author":"A Vaswani","year":"2017","unstructured":"Vaswani, A. et al. Attention is all you need. Adv. Neural Inform. Process. Sys. (NeurIPS) 30, 5998\u20136008 (2017).","journal-title":"Adv. Neural Inform. Process. Sys. (NeurIPS)"},{"key":"59312_CR28","unstructured":"Kenton, J. D. M.-W. C., Toutanova, L. K. Multi-head attention mechanism in Transformer architectures. In Proceedings of the conference of the North American chapter of the association for computational linguistics (NAACL), (2019)."},{"key":"59312_CR29","unstructured":"Michel, P., Levy, O., Neubig, G. Are sixteen heads really better than one? In Advances in Neural Information Processing Systems (NeurIPS), vol. 32, (2019)."},{"key":"59312_CR30","unstructured":"Rao, Y., Zhao, W., Liu, B., Lu, J., Zhou, J., Hsieh, C.-J. DynamicViT: Efficient vision Transformers with dynamic token sparsification. In Advances in neural information processing systems (NeurIPS), vol. 34, (2021)."},{"key":"59312_CR31","unstructured":"Romero, A., Ballas, N., Kahou, S. E., Chassang, A., Gatta, C., Bengio, Y. FitNets: Hints for thin deep nets. In Proceedings of the international conference on learning representations (ICLR), (2015)."},{"key":"59312_CR32","doi-asserted-by":"crossref","unstructured":"Yim, J., Joo, D., Bae, J., Kim, J. A gift from knowledge distillation: Fast optimization, network minimization and transfer learning. In Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition (CVPR), pp. 4133\u20134141, (2017).","DOI":"10.1109\/CVPR.2017.754"},{"key":"59312_CR33","doi-asserted-by":"crossref","unstructured":"Voita, E., Talbot, D., Moiseev, F., Sennrich, R., Titov, I. Analyzing multi-head self-attention: Specialized heads do the heavy lifting, the rest can be pruned. In Proceedings of the 57th annual meeting of the association for computational linguistics (ACL), pp. 5797\u20135808, (2019).","DOI":"10.18653\/v1\/P19-1580"},{"key":"59312_CR34","doi-asserted-by":"crossref","unstructured":"Kwon, W., Kim, S., Mahoney, M. W., Hassoun, J., Keutzer, K., Gholami, A. A fast post-training pruning framework for Transformers. In Advances in neural information processing systems (NeurIPS), vol. 35, (2022).","DOI":"10.52202\/068431-1750"},{"key":"59312_CR35","first-page":"24","volume":"505","author":"S Chen","year":"2022","unstructured":"Chen, S., Liu, Z., Jia, H., Li, Z. & Yan, S. Joint pruning and knowledge distillation for efficient neural network compression. Neurocomputing 505, 24\u201335 (2022).","journal-title":"Neurocomputing"},{"key":"59312_CR36","unstructured":"Jang, E., Gu, S., Poole, B. Categorical reparameterization with Gumbel-Softmax. In Proceedings of the international conference on learning representations (ICLR), (2017)."},{"key":"59312_CR37","unstructured":"Zhu, M., Tang, S., Zhang, K. Vision transformer pruning with sparsity regularization. In Proceedings of the IEEE\/CVF international conference on computer vision (ICCV), (2021)."},{"key":"59312_CR38","unstructured":"Nagel, Y., Fournarakis, M., Amjad, R. A., Bondarenko, Y., van Baalen, M., Blankevoort, T. A white paper on neural network quantization. Mach. Learn, (2021)."},{"key":"59312_CR39","doi-asserted-by":"crossref","unstructured":"Jacob, B., Kligys, S., Chen, B., Zhu, M., Tang, M., Howard, A., Adam, H., Kalenichenko, D. Quantization and training of neural networks for efficient integer-arithmetic-only inference. In Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition (CVPR), pp. 2704\u20132713, (2018).","DOI":"10.1109\/CVPR.2018.00286"},{"key":"59312_CR40","doi-asserted-by":"crossref","unstructured":"Yuan, Z., Xue, C., Chen, Y., Wu, Q., Sun, G. PTQ4ViT: Post-training quantization for vision Transformers with twin uniform quantization. In Proceedings of the European conference on computer vision (ECCV), pp. 191\u2013207, (2022).","DOI":"10.1007\/978-3-031-19775-8_12"},{"issue":"5","key":"59312_CR41","first-page":"1524","volume":"42","author":"J Li","year":"2023","unstructured":"Li, J., Wang, Y. & Zhang, T. Deploying neural networks on edge devices: Challenges and solutions for NPU-based inference. IEEE Trans. Comput. Aided Des. Integr. Circuits Syst. 42(5), 1524\u20131537 (2023).","journal-title":"IEEE Trans. Comput. Aided Des. Integr. Circuits Syst."},{"issue":"6","key":"59312_CR42","doi-asserted-by":"publisher","first-page":"1452","DOI":"10.1109\/TPAMI.2017.2723009","volume":"40","author":"B Zhou","year":"2018","unstructured":"Zhou, B., Lapedriza, A., Khosla, A., Oliva, A. & Torralba, A. Places: A 10 million image database for scene recognition. IEEE Trans. Pattern Anal. Mach. Intell. 40(6), 1452\u20131464 (2018).","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"issue":"12","key":"59312_CR43","doi-asserted-by":"publisher","first-page":"2341","DOI":"10.1109\/TPAMI.2010.168","volume":"33","author":"K He","year":"2011","unstructured":"He, K., Sun, J. & Tang, X. Single image haze removal using dark channel prior. IEEE Trans. Pattern Anal. Mach. Intell. 33(12), 2341\u20132353 (2011).","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"59312_CR44","doi-asserted-by":"publisher","first-page":"973","DOI":"10.1007\/s11263-018-1072-8","volume":"126","author":"C Sakaridis","year":"2018","unstructured":"Sakaridis, C., Dai, D. & Van Gool, L. Semantic foggy scene understanding with synthetic data. Int. J. Comput. Vision 126, 973\u2013992 (2018).","journal-title":"Int. J. Comput. Vision"},{"key":"59312_CR45","unstructured":"Katharopoulos, A., Vyas, A., Pappas, N., Fleuret, F. Transformers are RNNs: Fast autoregressive transformers with linear attention. In Proceedings of the 37th international conference on machine learning (ICML), vol. 119, pp. 5156\u20135165, (2020)."},{"key":"59312_CR46","doi-asserted-by":"crossref","unstructured":"Liu, Z., Mao, H., Wu, C.-Y., Feichtenhofer, C., Darrell, T., Xie, S. A ConvNet for the 2020s. In Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition (CVPR), pp. 11976\u201311986, (2022).","DOI":"10.1109\/CVPR52688.2022.01167"},{"key":"59312_CR47","doi-asserted-by":"crossref","unstructured":"Woo, S., Park, J., Lee, J.-Y., Kweon, I. S. CBAM: Convolutional block attention module. In Proceedings of the European conference on computer vision (ECCV), pp. 3\u201319, (2018).","DOI":"10.1007\/978-3-030-01234-2_1"},{"key":"59312_CR48","unstructured":"Loshchilov, I., Hutter, F. SGDR: Stochastic gradient descent with warm restarts. In Proceedings of the international conference on learning representations (ICLR), (2017)."},{"key":"59312_CR49","first-page":"15834","volume":"33","author":"T Chen","year":"2020","unstructured":"Chen, T. et al. The lottery ticket hypothesis for pre-trained BERT networks. Adv. Neural Inform. Process. Sys. (NeurIPS) 33, 15834\u201315846 (2020).","journal-title":"Adv. Neural Inform. Process. Sys. (NeurIPS)"},{"key":"59312_CR50","unstructured":"Loshchilov, I., Hutter, F. Decoupled weight decay regularization,\u201d In Proceedings of the international conference on learning representations (ICLR), (2019)."},{"key":"59312_CR51","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., Sun, J. Deep residual learning for image recognition. In Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition (CVPR), pp. 770\u2013778, (2016).","DOI":"10.1109\/CVPR.2016.90"},{"key":"59312_CR52","doi-asserted-by":"crossref","unstructured":"Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., Le, Q. V., Adam, H. Searching for MobileNetV3. In Proceedings of the IEEE\/CVF international conference on computer vision (ICCV), pp. 1314\u20131324, (2019).","DOI":"10.1109\/ICCV.2019.00140"}],"container-title":["Scientific Reports"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-59312-1","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-59312-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-59312-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,5]],"date-time":"2026-10-05T11:02:55Z","timestamp":1791198175000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41598-026-59312-1"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,4]]},"references-count":52,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["59312"],"URL":"https:\/\/doi.org\/10.1038\/s41598-026-59312-1","relation":{},"ISSN":["2045-2322"],"issn-type":[{"value":"2045-2322","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,4]]},"assertion":[{"value":"3 April 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 June 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 July 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"All authors have reviewed the manuscript and consent to its publication. No identifiable information regarding participants has been included.","order":2,"name":"Ethics","label":"Consent for publication","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study was approved by the Research Ethics Committee of Zhejiang Guangsha Vocational and Technical University of Construction (Reference Number: IRB-2024-03176). All participants involved in the data annotation process provided written informed consent prior to enrollment. The study was conducted in accordance with the Declaration of Helsinki and relevant national regulations, including the Personal Information Protection Law of the People\u2019s Republic of China. All images in the dataset were captured at publicly accessible coastal sites; identifiable faces and license plates were blurred before any analysis or release.","order":3,"name":"Ethics","label":"Ethics approval and consent to participate","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"30747"}}