基于ACE-YOLOv10的无人机船舶检测算法  

Drone ship detection algorithm based on ACE-YOLOv10

作　　者：刘畅[1] LIU Chang(College of Computer Science,Xi′an Shiyou University,Xi′an 710065,China)

机构地区：[1]西安石油大学计算机学院,西安710065

出　　处：《智能计算机与应用》2025年第3期152-157,共6页Intelligent Computer and Applications

摘　　要：无人机由于其成本低廉、敏捷、灵活并且可以搭载高分辨率摄像头和传感器,正在成为各个行业不可或缺的工具。在检测环境中,无人机可以适应复杂地形以及恶劣气候,收集到大气、土壤等各项珍贵数据。在工业探索领域,无人机可以代替人工对于危险地域的探查,降低事故的发生率。在海上救援方面,无人机可以快速定位到遇险的船舶,提升救援人员的救援效率。然而,无人机在此情况下的目标检测任务仍然面临不小挑战,如目标被障碍物遮挡、太阳光照的变化等。为了提高无人机在海上救援中的船舶检测性能,本文提出了一种基于改进YOLOv10(You Only Look Once,YOLO)的目标检测算法,加入了CA注意力机制,并采取EIoU损失函数。实验结果表明,相对于原始的YOLOv10模型,实验数据集在ACE-YOLOv10模型中获得了更好的结果。Due to its low cost,agility,flexibility,and the ability to carry high-resolution cameras and sensors,drones are increasingly becoming indispensable tools in various industries.In environmental detection,drones can adapt to complex terrains and harsh climates,and collect valuable data such as the atmosphere and soil.In the field of industrial exploration,drones can replace manual exploration of dangerous areas and reduce the incidence of accidents.In terms of maritime rescue,drones can be utilized to swiftly locate ships in distress,enhancing the rescue efficiency of rescue teams.However,the target detection tasks for drones in such scenarios still face considerable challenges,such as targets being obscured by obstacles,variations in sunlight,etc.To enhance the detection capabilities of drones in dealing with complex situations during maritime rescues,this paper proposes a target detection algorithm based on the improved YOLOv10(You Only Look Once,YOLO),adding the CA attention mechanism and adopting the EIoU loss function.Experimental results show that compared with the original YOLOv10 model,the experimental dataset obtains better results in the ACE-YOLOv10 model.

关 键 词：无人机 YOLOv10 目标检测 CA注意力机制 EIoU损失函数 海上救援 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]