基于改进YOLOv8n的船舶设备拆装流程规范性评估方法  

Operation standardization evaluation method based on improved YOLOv8n for ship equipment disassembly and assembly

作　　者：张振东 管聪 张泽辉 吴超[1] 丁学文 ZHANG Zhendong;GUAN Cong;ZHANG Zehui;WU Chao;DING Xuewen(School of Naval Architecture,Ocean and Energy Power Engineering,Wuhan University of Technology,Wuhan 430063,China;School of Automation(School of Artificial Intelligence),Hangzhou Dianzi University,Hangzhou 310018,China)

机构地区：[1]武汉理工大学船海与能源动力工程学院,湖北武汉430063 [2]杭州电子科技大学自动化学院(人工智能学院),浙江杭州310018

出　　处：《中国舰船研究》2025年第2期140-150,共11页Chinese Journal of Ship Research

基　　金：浙江省基础公益研究计划项目(LTGG24F030004);国家水运安全工程技术研究中心开放基金资助项目(A202403);国家重点研发计划项目(2022YFE0210700);中央高校基本科研业务费专项资金资助项目(104972024JYS0043)。

摘　　要：[目的]船舶机舱作业规范性是船舶安全管控的关键部分,因此船员实操考试将船舶设备拆装作为一个重要考核项。为提升船员实操考试的电子化和智能化水平,提出一种基于计算机视觉的船舶设备拆装流程规范性的自动化识别方法。[方法]首先,以YOLOv8n构建船舶设备检测模型的骨干网络,并引入高效通道注意力机制(SA),以提高模型特征提取能力与训练效率;然后,在颈部网络中引入重参数化泛化特征的金字塔网络(GFPN)融合结构,以提高模型的多尺度特征融合能力;最后,引入动态非单调聚焦机制损失函数(WIoU)来替换原CIoU损失函数,以提高模型精度。[结果]自建数据集的试验结果表明:与YOLOv8n相比,改进目标识别算法的平均精度均值提高了0.15,实时检测的每秒帧数提升了0.6,可以准确识别齿轮泵的拆装流程。[结论]该改进算法具有更强的识别能力,可以更好地应用于船舶设备拆装流程规范性的识别任务。[Objectives]The standardization of ship engine room operations is a critical component of ship safety management.Therefore,the practical examination for crew members includes the disassembly and assembly of ship equipment as a key assessment item.To enhance the digitalization and intelligence of crew practical examinations,a computer vision-based automated recognition method for assessing the standardization of ship equipment disassembly and assembly processes is proposed.[Methods]First,the backbone network of the ship equipment detection model is constructed using YOLOv8n,and the shuffle-attention(SA)mechanism is introduced to improve the model's feature extraction capability and training efficiency.Subsequently,a reparameterized generalized feature pyramid network(GFPN)fusion structure is incorporated into the neck network to enhance the model's ability to fuse multi-scale features.Finally,the original CIoU loss function is replaced with the wise intersection over union(WIoU)loss function to improve the model's accuracy.[Results]Experimental results on a self-constructed dataset demonstrate that,compared to YOLOv8n,the improved object detection algorithm achieves a 0.15 increase in mean average precision and a 0.6 framesper-second improvement in real-time detection,enabling accurate recognition of the disassembly and assembly processes of gear pumps.[Conclusion]The improved algorithm exhibits superior recognition capabilities and is better suited for identifying the standardization of ship equipment disassembly and assembly processes.

关 键 词：船舶设备 拆除和安装 目标检测 注意力机制(SA) 泛化特征金字塔网络(GFPN) 动态非单调聚焦机制(WIoU)损失函数 

分 类 号：U676.2[交通运输工程—船舶及航道工程]