计算机工程与应用 ›› 2023, Vol. 59 ›› Issue (12): 201-207.DOI: 10.3778/j.issn.1002-8331.2302-0157

• 图形图像处理 • 上一篇    下一篇

改进YOLOv5的小目标检测算法

俞军,贾银山   

  1. 辽宁石油化工大学 人工智能与软件学院,辽宁 抚顺 113000
  • 出版日期:2023-06-15 发布日期:2023-06-15

Improved YOLOv5 for Small Object Detection Algorithm

YU Jun, JIA Yinshan   

  1. College of Artificial Intelligence and Software, Liaoning Petrochemical University, Fushun, Liaoning 113000, China
  • Online:2023-06-15 Published:2023-06-15

摘要: 虽然现在的深度学习技术在大中目标检测领域取得了惊人的进步,但是由于小目标的尺寸有限以及卷积网络的局限性,导致小目标检测仍然是一个具有挑战性的问题。通过改进YOLOv5算法,提出了一种针对小目标的YOLO-S模型。在原来三层输出层的基础上,利用级联网络,添加一个专门针对于小目标检测的输出层。为了补充上下文信息以及抑制多尺度特征融合冲突,设计了一种新的上下文信息提取模块CFM(Context Feature Module)以及基于通道和空间特征细化的模块FSM(feature specify module)。上采样方式由原来的最邻近插值替换为新设计的Transpose模块,可以将信息最大化恢复。数据集采用专门针对于小目标的VisDrone2019来验证算法的有效性。实验结果表明,YOLO-S比YOLOv5的mAP@0.5提高了6.9个百分点。

关键词: YOLOv5, 小目标检测, 级联网络, 上下文信息, 特征细化

Abstract: Although the current deep learning technology has made amazing progress in the field of large and medium object detection, small object detection is still a challenging problem today due to the limited size of small object and the limitations of convolutional networks. Based on You Only Look Once version 5(hereinafter referred to as YOLOv5) algorithm, this research proposes a YOLO-S model, which is very friendly to small objects. Firstly, on the basis of the orginal output layer with only three layers, a special output layer for small object detection is added by using the cascade network. Secondly, in order to supplement context information and suppress multi-scale feature fusion conflicts, a new supplement context information module CFM and channel and spatial feature thinning module FSM is designed. Finally, the upsampling method is replaced by deconvolution from the original linear interpolation. The dataset uses VisDrone2019, which is specially designed for small objects, to verify the effectiveness of the algorithm. The experimental results show that the mAP@0.5 of YOLO-S is 6.9 percentage points higher than that of YOLOv5.

Key words: You Only Look Once version 5(YOLOv5), small object, cascade network, context information, feature refinement