Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (11): 3603-3609.DOI: 10.11772/j.issn.1001-9081.2023111644

• Multimedia computing and computer simulation • Previous Articles     Next Articles

Small object detection algorithm from drone perspective based on improved YOLOv8n

Tao LIU1,2, Shihong JU1(), Yimeng GAO1   

  1. 1.School of Software,Liaoning Technical University,Huludao Liaoning 125105,China
    2.Department of Basic Teaching,Liaoning Technical University,Huludao Liaoning 125105,China
  • Received:2023-12-01 Revised:2024-04-05 Accepted:2024-04-12 Online:2024-05-30 Published:2024-11-10
  • Contact: Shihong JU
  • About author:LIU Tao, born in 1981, M. S., associate professor. His research interests include computer vision, intelligent data processing.
    GAO Yimeng, born in 1996, M. S. candidate. Her research interests include image processing, pattern recognition, artificial intelligence.
  • Supported by:
    National Natural Science Foundation of China(61172144)

基于改进YOLOv8n的无人机视角下小目标检测算法

刘涛1,2, 鞠事宏1(), 高一萌1   

  1. 1.辽宁工程技术大学 软件学院,辽宁 葫芦岛 125105
    2.辽宁工程技术大学 基础教学部,辽宁 葫芦岛 125105
  • 通讯作者: 鞠事宏
  • 作者简介:刘涛(1981—),男,河北定州人,副教授,硕士,主要研究方向:计算机视觉、智能数据处理
    高一萌(1996—),女,辽宁鞍山人,硕士研究生,主要研究方向:图像处理、模式识别、人工智能。
  • 基金资助:
    国家自然科学基金资助项目(61172144)

Abstract:

In view of the low accuracy of object detection algorithms in small object detection from drone perspective, a new small object detection algorithm named SFM-YOLOv8 was proposed by improving the backbone network and attention mechanism of YOLOv8. Firstly, the SPace-to-Depth Convolution (SPDConv) suitable for low-resolution images and small object detection was integrated into the backbone network to retain discriminative feature information and improve the perception ability to small objects. Secondly, a multi-branch attention named MCA (Multiple Coordinate Attention) was introduced to enhance the spatial and channel information on the feature layer. Then, a convolution FE-C2f fusing FasterNet and Efficient Multi-scale Attention (EMA) was constructed to reduce the computational cost and lightweight the model. Besides, a Minimum Point Distance based Intersection over Union (MPDIoU) loss function was introduced to improve the accuracy of the algorithm. Finally, a small object detection layer was added to the network structure of YOLOv8n to retain more location information and detailed features of small objects. Experimental results show that compared with YOLOv8n, SFM-YOLOv8 achieves a 4.37 percentage point increase in mAP50 (mean Average Precision) with a 5.98% reduction in parameters on VisDrone-DET2019 dataset. Compared to the related mainstream models, SFM-YOLOv8 achieves higher accuracy and meets real-time detection requirements.

Key words: small object detection, YOLOv8, feature extraction, attention mechanism, loss function

摘要:

针对目标检测算法在无人机视角下的小目标检测中精度低的问题,通过改进YOLOv8的骨干网络与注意力机制,提出一种新的小目标检测算法SFM-YOLOv8。首先,在骨干网络中融入适用于低分辨率图像和小物体检测的空间深度转换卷积(SPDConv),保留判别特征信息,提高小目标感知能力;其次,插入多分支注意力MCA(Multiple Coordinate Attention),加强提取特征层的空间信息和通道信息;然后,构建一种融合FasterNet和高效多尺度注意力(EMA)的卷积FE-C2f,减少计算量并使模型轻量化;此外,引入边界框相似度比较度量(MPDIoU)损失函数提高算法精度;最后,在YOLOv8n的网络结构中增加小目标检测层,保留更多关于小目标的位置信息和细节特征。实验结果表明,与YOLOv8n相比,SFM-YOLOv8算法在VisDrone-DET2019数据集上的平均精度均值mAP50提高了4.37个百分点,参数量减少了5.98%;与相关主流模型对比,精度也有所提升,且满足实时检测需求。

关键词: 小目标检测, YOLOv8, 特征提取, 注意力机制, 损失函数

CLC Number: