基于改进YOLOv8的轻量化车辆检测网络

doi:10.19678/j.issn.1000-3428.0069122

摘要/Abstract

摘要： 现有的高精度车辆检测模型参数与计算量过高,无法在交通智能设备上良好运行,而轻量化的车辆检测模型精度普遍较低,不适用于实际任务。为此,提出一种改进YOLOv8的轻量化车辆检测网络,将主干网络替换为计算量和内存访问更小的FasterNet网络,并且将颈部的双向特征金字塔网络替换为加权双向特征金字塔网络(BiFPN),简化特征融合过程。同时,引入一种融合注意力机制的动态检测头,实现检测头和注意力的无冗余结合;此外,针对完全交并比(CIoU)在检测精度和收敛速度上的缺陷,提出一种尺度不变交并比(SIoU)结合归一化高斯Wasserstein距离(NWD)的回归损失算法。最后,为尽量减小模型对边缘设备的算力需求,进行基于幅值的层自适应稀疏化剪枝,进一步压缩模型大小。实验结果表明,提出的改进模型相较于原模型YOLOv8s,在精度上升1.5百分点的情况下,参数量降低78.9%,计算量下降67.4%,模型尺寸降低77.8%,达到了比较优秀的轻量化效果,具有很强的实用性。

关键词: YOLOv8模型, 车辆检测, 轻量化, FasterNet网络, 归一化高斯Wasserstein距离

Abstract: The current high-precision vehicle detection model faces challenges due to its excessive parameterization and computational demands, making it unsuitable for efficient operation on intelligent transportation devices. Conversely, lightweight vehicle detection models often sacrifice accuracy, rendering them unsuitable for practical tasks. In response, an improved lightweight vehicle detection network based on YOLOv8 is proposed. This enhancement involves substituting the main network with the FasterNet architecture, which reduces the computational and memory access requirements. Additionally, we replace the Bidirectional Feature Pyramid Network (BiFPN) in the neck with a weighted bidirectional feature pyramid network to simplify the feature fusion process. Simultaneously, we introduce a dynamic detection head with a fusion attention mechanism to achieve nonredundant integration of the detection head and attention. Furthermore, we address the deficiencies of the Complete Intersection over Union (CIoU) in terms of detection accuracy and convergence speed by proposing a regression loss algorithm that incorporates the Scale-invariant Intersection over Union (SIoU) combined with the Normalized Gaussian Wasserstein Distance (NWD). Finally, to minimize the computational demands on edge devices, we implement amplitude-based layer-wise adaptive sparsity pruning, which further compresses the model size. Experimental results demonstrate that the proposed improved model, compared with the original YOLOv8s model, achieves a 1.5 percentage points increase in accuracy, a 78.9% reduction in parameter count, a 67.4% decrease in computational demands, and a 77.8% reduction in model size. This demonstrates the outstanding lightweight effectiveness and practical utility of the proposed model.

Key words: YOLOv8 model, vehicle detection, lightweight, FasterNet network, Normalized Gaussian Wasserstein Distance(NWD)

中图分类号:

TP391

陈梓延, 王晓龙, 何迪, 安国成. 基于改进YOLOv8的轻量化车辆检测网络[J]. 计算机工程, 2025, 51(5): 314-325.

CHEN Ziyan, WANG Xiaolong, HE Di, AN Guocheng. Lightweight Vehicle Detection Network Based on Improved YOLOv8[J]. Computer Engineering, 2025, 51(5): 314-325.

https://www.ecice06.com/CN/Y2025/V51/I5/314

参考文献

[1] 李国进,胡洁,艾矫燕.基于改进SSD算法的车辆检测[J].计算机工程, 2022, 48(1):266-274. LI G J, HU J, AI J Y. Vehicle detection based on improved SSD algorithm[J]. Computer Engineering, 2022, 48(1):266-274.(in Chinese)
[2] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once:unified, real-time object detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2016:335-347.
[3] BERG A C, FU C Y, SZEGEDY C, et al. SSD:single shot MultiBox detector[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany:Springer, 2015:557-565.
[4] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2):318-327.
[5] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of 2014 IEEE Conference on Computer Vision and Pattern Recognition. New York, USA:ACM Press, 2014:580-587.
[6] HE K M, ZHANG X Y, REN S Q, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[M]. Berlin, Germany:Springer, 2014.
[7] REN S, HE K, GIRSHICK R, et al. Faster R-CNN:towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6):1137-1149.
[8] 郭克友,王苏东,李雪,等.基于Dim env-YOLO算法的昏暗场景车辆多目标检测[J].计算机工程, 2023, 49(3):312-320. GUO K Y, WANG S D, LI X, et al. Multi-target detection of vehicles in dim scenes based on Dim env-YOLO algorithm[J]. Computer Engineering, 2023, 49(3):312-320.(in Chinese)
[9] BOCHKOVSKIY A, WANG C Y, LIAO H. YOLOv4:optimal speed and accuracy of object detection[EB/OL].[2023-11-20] . https://arxiv.org/abs/2004.10934.pdf.
[10] HOWARD A, SANDLER M, CHU G, et al. Searching for MobileNetV3[EB/OL].[2023-11-20] . https://arxiv.org/abs/1905.02244.pdf.
[11] 李松江,耿兰兰,王鹏.基于改进Yolov4的车辆目标检测[J].计算机工程, 2023, 49(4):272-280. LI S J, GENG L L, WANG P. Vehicle target detection based on improved Yolov4[J]. Computer Engineering, 2023, 49(4):272-280.(in Chinese)
[12] WANG Q, WU B, ZHU P, et al. ECA-Net:efficient channel attention for deep convolutional neural networks[EB/OL].[2023-11-20] . http://arxiv.org/pdf/1910.03151v4.pdf.
[13] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2015:1-9.
[14] 郑秋梅,王璐璐,王风华.基于改进卷积神经网络的交通场景小目标检测[J].计算机工程, 2020, 46(6):26-33. ZHENG Q M, WANG L L, WANG H. Small object detection in traffic scene based on improved convolutional neural network[J]. Computer Engineering, 2020, 46(6):26-33.(in Chinese)
[15] REDMON J, FARHADI A. YOLOv3:an incremental improvement[EB/OL].[2023-11-20] . https://arxiv.org/pdf/1804.02767.
[16] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2016:770-778.
[17] LIAN J, YIN Y, LI L, et al. Small object detection in traffic scenes based on attention feature fusion[J]. Sensors, 2021, 21(9):3031.
[18] CAI Y, LUAN T, GAO H, et al. YOLOv4-5D:an effective and efficient object detector for autonomous driving[J]. IEEE Transactions on Instrumentation and Measurement, 2021,70(1):1-13.
[19] WANG R, WANG Z, XU Z, et al. A real-time object detector for autonomous vehicles based on YOLOv4[J]. Computational Intelligence and Neuroscience, 2021, 2021:1.
[20] 原蕾,王科俊.基于注意力机制与特征融合的改进YOLOv7车辆检测方法[J].国外电子测量技术, 2023, 42(9):49-57. YUAN L, WANG K J. Vehicle detection based on YOLOv7 improved by attention mechanism and feature fusion[J]. Foreign Electronic Measurement Technology, 2023, 42(9):49-57.(in Chinese)
[21] 蔡刘畅,杨培峰,张秋仪.基于YOLOv7的道路监控车辆检测方法[J].陕西科技大学学报, 2023, 41(6):155-161, 175. CAI L C, YANG P F, ZHANG Q Y. Vehicle detection method based on YOLOv7 in traffic monitoring[J]. Journal of Shaanxi University of Science& Technology, 2023, 41(6):155-161, 175.(in Chinese)
[22] WANG C Y, BOCHKOVSKIY A, LIAO H M. YOLOv7:trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2023:267-275.
[23] HAN K, WANG Y, TIAN Q, et al. GhostNet:more features from cheap operations[EB/OL].[2023-11-20] . https://arxiv.org/pdf/1911.11907.pdf.
[24] 宋华杰,周磊.基于函数改进的YOLOv3车辆检测与识别算法研究[J].智能科学与技术学报, 2023, 5(4):535-542. SONG H J, ZHO L. Vehicle detection and recognition algorithm based on function improvement of YOLOv3[J]. Chinese Journal of Intelligent Science and Technology, 2023, 5(4):535-542.(in Chinese)
[25] 郭奕裕,周箩鱼.安全帽佩戴检测网络模型的轻量化设计[J].计算机工程, 2023, 49(4):312-320. GUO Y Y, ZHOU L Y. Lightweight design of safety helmet wearing detection network model[J]. Computer Engineering, 2023, 49(4):312-320.(in Chinese)
[26] LIU S, QI L, QIN H, et al. Path aggregation network for instance segmentation[C]// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2018:8759-8768.
[27] LI X, WANG W, WU L, et al. Generalized focal loss:learning qualified and distributed bounding boxes for dense object detection[EB/OL].[2023-11-20] . http://arxiv.org/abs/2006.04388.
[28] CHEN J, KAO S, HE H, et al. Run, don't walk:chasing higher FLOPS for faster neural networks[C]//Proceedings of 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2023:365-376.
[29] TAN M X, PANG R M, LE Q V. EfficientDet:scalable and efficient object detection[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2020:2535-2546.
[30] DAI X Y, CHEN Y P, XIAO B, et al. Dynamic head:unifying object detection heads with attentions[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2021:479-488.
[31] GEVORGYAN Z. SIoULoss:more powerful learning for bounding box regression[EB/OL].[2023-11-20] . https://arxiv.org/abs/2205.12740.
[32] WANG J, XU C, YANG W, et al. A normalized gaussian wasserstein distance for tiny object detection[EB/OL].[2023-11-20] . https://arxiv.org/abs/2110.13389.
[33] ZHANG X Y, ZHOU X Y, LIN M X, et al. ShuffleNet:an extremely efficient convolutional neural network for mobile devices[C]// Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2018:354-365.
[34] CHOLLET F. Xception:deep learning with depthwise separable convolutions[C]// Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2017:1800-1807.
[35] LI H, LI J, WEI H, et al. Slim-neck by GSConv:a better design paradigm of detector architectures for autonomous vehicles[EB/OL].[2023-11-20] . https://arxiv.org/abs/2206.02424.
[36] LEE J, PARK S, MO S, et al. Layer-adaptive sparsity for the magnitude-based pruning[EB/OL].[2023-11-20] . https://arxiv.org/abs/2010.07611v2.
[37] YANG G Y, LEI J, ZHU Z K, et al. AFPN:asymptotic feature pyramid network for object detection[C]//Proceedings of IEEE International Conference on Systems, Man, and Cybernetics. Washington D. C., USA:IEEE Press, 2023:663-675.
[38] REZATOFIGHI H, TSOI N, GWAK J, et al. Generalized intersection over union:a metric and a loss for bounding box regression[C]// Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2019:483-492.
[39] ZHANG Y F, REN W Q, ZHANG Z, et al. Focal and efficient IOU loss for accurate bounding box regression[J]. Neurocomputing, 2022, 506:146-157.
[40] ZHANG H, XU C, ZHANG S. Inner-IoU:more effective intersection over union loss with auxiliary bounding box[EB/OL].[2023-11-20] . https://arxiv.org/abs/2311.02877.
[41] MA S L, XU Y. MPDIoU:a loss for efficient and accurate bounding box regression[EB/OL].[2023-11-20] . https://arxiv.org/abs/2307.07662.
[42] LECUN Y, DENKER J, SOLLA S. Optimal brain damage[C]//Proceedings of Advances in Neural Information Processing Systems. Cambridge, USA:MIT Press, 1989:667-676.
[43] LI H, KADAV A, DURDANOVIC I, et al. Pruning filters for efficient ConvNets[EB/OL].[2023-11-20] . https://arxiv.org/pdf/1608.08710.
[44] MOLCHANOV P, MALLYA A, TYREE S, et al. Importance estimation for neural network pruning[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2019:11256-11264.
[45] FANG G F, MA X Y, SONG M L, et al. DepGraph:towards any structural pruning[EB/OL].[2023-11-20] . https://arxiv.org/abs/2301.12900.

选择文件类型/文献管理软件名称

选择包含的内容