Computer Engineering and Applications ›› 2024, Vol. 60 ›› Issue (17): 191-202.DOI: 10.3778/j.issn.1002-8331.2403-0256

• Graphics and Image Processing • Previous Articles     Next Articles

AEM-YOLOv8s:Small Target Detection Algorithm for UAV Aerial Images

JIANG Wei, WANG Wanhu, YANG Junjie   

  1. 1.College of Electronics and Information Engineering, Shanghai University of Electric Power, Shanghai 201306, China
    2.Shanghai Dianji University, Shanghai 201306, China
  • Online:2024-09-01 Published:2024-08-30

AEM-YOLOv8s:无人机航拍图像的小目标检测

蒋伟,王万虎,杨俊杰   

  1. 1.上海电力大学 电子与信息工程学院,上海 201306
    2.上海电机学院,上海 201306

Abstract: The AEM-YOLOv8s algorithm is proposed to address issues of low performance, missed detections, occlusions, and high model parameter count in small object detection in current UAV aerial imagery. Within the C2f module, the advantages of AKConv (alterable kernel convolution) and EMA (efficient multi-scale attention) are combined to design the C2f-BE module, which enhances the algorithm’s ability to process features while reducing the model parameter count. By introducing a small object detection layer and BiFPN structure, through cross-scale connections and weighted feature fusion, more shallow features are retained, reducing algorithm parameters. The design of a multi-scale feature fusion branch merges shallow features containing more small object information with deeper semantic features, reducing missed detections under occlusion and improving small object detection performance. Experimental results on the VisDrone2019 public dataset demonstrate that the AEM-YOLOv8s algorithm achieves an mAP50 of 50.1% and mAP50:95 of 31.1%, representing respective improvements of 10.8 and 7.6 percentage points over YOLOv8s, while also reducing parameters by 32.2% compared to YOLOv8s.

Key words: YOLOv8s, C2f-BE module, small object, multi-scale

摘要: 针对目前无人机航拍图中的小目标检测性能低、漏检、遮挡以及模型参数量大的问题,提出了AEM-YOLOv8s算法。在C2f模块中结合AKConv(alterable kernel convolution)和EMA(efficient multi-scale attention)的优点,设计了C2f-BE模块,更好地提高了算法处理特征的能力,同时也降低了模型参数量。引入小目标检测层和BiFPN结构,通过跨尺度连接方式和加权特征融合,能够保留更多的浅层特征,并且减少了算法参数量。设计多尺度特征融合分支,将浅层特征与深层特征进行融合,减少了遮挡情况下的漏检,提高了算法对小目标检测性能。在VisDrone2019公开数据集上的实验表明,AEM-YOLOv8s算法的mAP50为50.1%,mAP50:95为31.1%,较YOLOv8s分别提高了10.8和7.6个百分点,同时参数量较YOLOv8s降低了32.2%。

关键词: YOLOv8s, C2f-BE模块, 小目标, 多尺度