eprintid: 10199771 rev_number: 7 eprint_status: archive userid: 699 dir: disk0/10/19/97/71 datestamp: 2024-11-08 08:42:11 lastmod: 2024-11-08 08:42:11 status_changed: 2024-11-08 08:42:11 type: article metadata_visibility: show sword_depositor: 699 creators_name: Shao, Zeyuan creators_name: Yin, Yong creators_name: Lyu, Hongguang creators_name: Soares, C Guedes creators_name: Cheng, Tao creators_name: Jing, Qianfeng creators_name: Yang, Zhilin title: An efficient model for small object detection in the maritime environment ispublished: pub divisions: UCL divisions: B04 divisions: F44 keywords: Autonomous navigation, Sea surface, Object detection Deep learning, Model quantization note: This version is the author accepted manuscript. For information on re-use, please refer to the publisher's terms and conditions. abstract: Environmental perception is crucial for autonomous ships realizing autonomous navigation, in particular, the high-precision and low-latency detection of small objects on the sea surface is a key and challenging issue. To address this problem, this paper presents a model that improves the detection accuracy and delivers excellent real-time performance for autonomous ship navigation. The backbone of the proposed model enhances the modelling capabilities by expanding deformable convolution and introducing the self-designed attention mechanism. Additionally, an enhanced feature fusion structure is designed by the pixel shuffle based on super-resolution reconstruction to keep the integrity of feature information for small objects. This paper also presents an optimized model quantization strategy that alleviates the problem of low model efficiency caused by the limited resources onboard the ship. Compared to the earlier model, the present one has increased the mean average precision on the Rizhao Zhuimeng-3# maritime optical dataset by 4.5 % and by 21 % for small object detection. Furthermore, the real-time detection for high-resolution images can now reach a speed of 67 frames/s. Moreover, the present model outperforms existing methods concerning accuracy when the frames/s are similar. The results indicate that the proposed model can be potentially applied to autonomous ships. date: 2024-11 date_type: published publisher: Elsevier official_url: http://dx.doi.org/10.1016/j.apor.2024.104194 full_text_type: other language: eng verified: verified_manual elements_id: 2311536 doi: 10.1016/j.apor.2024.104194 lyricists_name: Cheng, Tao lyricists_id: TCHEN23 actors_name: Cheng, Tao actors_id: TCHEN23 actors_role: owner funding_acknowledgements: 2022YFB4300803 [National Key R & D Program of China]; 2022YFB4301402 [National Key R & D Program of China]; 52071049 [National Natural Science Foundation of China]; 2022JH1/10800096 [Liaoning Provincial Science and Technology Plan (Key) project]; CSC [2022] 2260 [Chinese Scholarships Council]; UIDB/UIDP/00134/2020 [Portuguese Foundation for Science and Technology (Fundaca o para a Ciencia e Tecnologia-FCT)] full_text_status: restricted publication: Applied Ocean Research volume: 152 article_number: 104194 pages: 16 citation: Shao, Zeyuan; Yin, Yong; Lyu, Hongguang; Soares, C Guedes; Cheng, Tao; Jing, Qianfeng; Yang, Zhilin; (2024) An efficient model for small object detection in the maritime environment. Applied Ocean Research , 152 , Article 104194. 10.1016/j.apor.2024.104194 <https://doi.org/10.1016/j.apor.2024.104194>. document_url: https://discovery.ucl.ac.uk/id/eprint/10199771/1/An%20Efficient%20Model%20for%20Small%20Object%20Detection%20in%20the%20Maritime%20Environment.pdf