面向复杂场景密集行人检测的YOLOv8改进模型  

Improved YOLOv8 model for dense pedestrian detection in complex scenes

作　　者：胡伟超 皮建勇[1] 胡倩 黄昆 王娟敏 Hu Weichao;Pi Jianyong;Hu Qian;Huang Kun;Wang Juanmin(State Key Laboratory of Public Big Data,College of Computer Science and Technology,Guizhou University,Guiyang 550025,China)

机构地区：[1]贵州大学计算机科学与技术学院公共大数据国家重点实验室,贵阳550025

出　　处：《电子测量技术》2024年第14期159-169,共11页Electronic Measurement Technology

基　　金：贵州省科技支撑计划(黔科合支撑[2023]一般430)项目资助。

摘　　要：针对当前行人检测面临的环境复杂、目标尺寸多变和严重遮挡等挑战,导致现有检测技术在识别密集行人时容易发生误判和遗漏的问题,本文提出一种高效的面向复杂场景密集行人检测的YOLOv8改进模型。在骨干网络引入DCNv2设计C2f_DCNetv2替换C2f模块,提升骨干网络的特征提取能力;通过在架构中加入小目标检测头,增强模型对小尺寸目标的检测能力,提高对小目标的检测识别精度;基于四检测头改进AFPN设计出AFPN-4H,优化特征层之间的信息融合,提高了模型对不同尺度目标的适应性和检测精度;最后,通过结合Wise-IoU、Focaler-IoU和MPDIoU得到WFM-IoU,进一步提高了目标定位的准确性。实验结果表明,与原始的YOLOv8n模型相比,在P、R、AP50以及AP50:95等关键指标上分别提升1.6%、4.0%、3.6%和3.8%,也优于其他算法。验证了本文改进算法在复杂场景密集行人检测任务中具有较好的性能。Aiming at the current challenges of pedestrian detection,such as complex environments,variable target sizes,and severe occlusions,which cause existing detection techniques to be prone to misjudgment and omission when recognising dense pedestrians,this paper proposes an efficient YOLOv8 improved model for dense pedestrian detection in complex scenes.DCNv2 is introduced into the backbone network,and C2f_DCNetv2 is designed to replace the C2f module,which improves the feature extraction capability of the backbone network;the detection capability of the model for small targets is improved by adding small-target detecting heads to the architecture,which improves the accuracy of small-target detection and recognition;based on the four detecting heads as well as the AFPN,the AFPN-4H is designed,which optimises the information fusion between the feature layers and improves the model's adaptability and detection accuracy for targets of different scales;finally,through the combination of Wise-IoU,Focaler-IoU,and MPDIoU,the WFM-IoU is obtained,which further improves the target localisation accuracy.The experimental results show that compared with the original YOLOv8n model,it improves 1.6,4.0,3.6 and 3.8 percentage points in the key indexes of P,R,AP50 and AP50:95,respectively,which are also inferior to other algorithms.The improved algorithm in this paper has better performance in the dense pedestrian detection task in complex scenes.

关 键 词：密集行人检测 YOLOv8 可变形卷积 多尺度特征融合 损失函数 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术] TN98[自动化与计算机技术—计算机科学与技术]