Computer Engineering and Applications ›› 2025, Vol. 61 ›› Issue (3): 111-120.DOI: 10.3778/j.issn.1002-8331.2407-0520

• Special Issue on YOLOv8 Improvements and Applications • Previous Articles     Next Articles

Small Target Detection Algorithm for UAV Based on Composite Feature and Multi-Scale Fusion

LIAO Ningsheng, CAO Tianxiu, LIU Keyan, XU Meng, ZHU Mi, GU Yuxuan, WANG Pengfei   

  1. 1.School of Artificial Intelligence, Chongqing University of Technology, Chongqing 401135, China
    2.Chongqing Construction Industry Co., Ltd., Chongqing 400054, China
  • Online:2025-02-01 Published:2025-01-24

复合特征与多尺度融合的无人机小目标检测算法

廖宁生,曹天秀,刘科言,徐猛,朱秘,古宇轩,王朋飞   

  1. 1.重庆理工大学 两江人工智能学院,重庆 401135
    2.重庆建设工业(集团)有限责任公司,重庆 400054

Abstract: A UAV target detection algorithm with composite feature and multi-scale fusion, CM-YOLOv8s(composite and multi-scale YOLOv8s) is proposed to address the issues of missed detections, false detections, and the imbalance between accuracy and speed in current UAV perspective detection algorithms. The quality of target composite features is improved by introducing channel features in the spatial pyramid pooling module. The neck structure of the model  is redesigned to improve the retention ratio of detailed target features. Additionally, the DRHead detection head is designed to achieve multi-scale feature map fusion, enhancing the adaptability for multi-scale target detection. The Wise-IoU loss function is employed to accelerate model convergence. Compared to the baseline algorithm, the improved CM-YOLOv8s algorithm has a parameter count of only 3.5×106, reducing the parameters by 69%. Experimental results show that the proposed CM-YOLOv8s algorithm significantly improves the mAP50 by 6.8 percentage poins on the VisDrone2019 dataset. Furthermore, the generalization and effectiveness of the proposed algorithm are validated on the UAV-DT and DIOR datasets.

Key words: unmanned aerial vehicle (UAV), small target, YOLOv8, lightweight, composite feature, multi-scale feature fusion

摘要: 针对目前无人机航拍图像小目标检测算法存在的漏检、误检、精度与速度不平衡等问题,提出复合特征与多尺度融合的无人机小目标检测算法CM-YOLOv8s(composite and multi-scale YOLOv8s)。通过在空间金字塔池化模块引入通道特征,实现目标复合特征质量提升;通过重建模型颈部结构,提高目标细节特征的保留比例;通过设计DRHead检测头,实现多尺度特征检测图融合,增强多尺度目标检测适应性;通过采用Wise-IoU损失函数提升模型收敛速度。相比于基准算法,改进后的CM-YOLOv8s算法参数量仅为3.5×106,参数量降低了69%。实验结果表明,提出的CM-YOLOv8s算法在数据集VisDrone2019上的mAP50显著提升了6.8个百分点;同时,在UAV-DT和DIOR数据集上验证了提出算法的泛化性和有效性。

关键词: 无人机(UAV), 小目标, YOLOv8, 轻量级, 复合特征, 多尺度特征融合