基于YOLOv8改进的跌倒检测算法: OEF-YOLO

doi:10.19678/j.issn.1000-3428.0069257

摘要/Abstract

摘要：

在室内场景下, 受角度、光线变化等因素的影响, 导致现有目标检测算法检测跌倒事件时检测精度降低、实时性变差。为此, 提出一种基于YOLOv8改进的跌倒检测算法OEF-YOLO。采用全维动态卷积(ODConv)模块改进YOLOv8中的C2f模块, 优化了核空间的4个维度以增强特征提取能力, 而且有效减少了计算负担。同时, 为了捕获更细粒度的特征, 在颈部网络中引入高效多尺度注意力(EMA)模块, 进一步聚合像素级特征, 提高网络在跌倒场景中的处理能力。在CIoU损失函数中融入Focal Loss思想, 使模型对难分类样本给予更多关注, 优化模型整体性能。实验结果表明, 相比YOLOv8n, OEF-YOLO跌倒检测算法在mAP@0.5指标上提升了1.5百分点, mAP@0.5∶0.95提升1.4百分点, 参数量和计算量分别为3.1×10⁶和6.5 GFLOPs, 在图形处理器(GPU)上FPS提高了44, 在提高精度检测跌倒事件的同时, 兼顾了低算力场景下的部署要求。

关键词: 目标检测, 轻量化, 跌倒事件, 注意力机制, 全维动态卷积

Abstract:

Existing object detection algorithms suffer from low detection accuracy and poor real-time performance when detecting fall events in indoor scenes, owing to changes in angle and light. In response to this challenge, this study proposes an improved fall detection algorithm based on YOLOv8, called OEF-YOLO. The C2f module in YOLOv8 is improved by using a Omni-dimensional Dynamic Convolution (ODConv) module, optimizing the four dimensions of the kernel space to enhance feature extraction capabilities and effectively reduce computational burden. Simultaneously, to capture finer grained features, the Efficient Multi-scale Attention (EMA) module is introduced into the neck network to further aggregate pixel-level features and improve the network's processing ability in fall scenes. Integrating the Focal Loss idea into the Complete Intersection over Union (CIoU) loss function allows the model to pay more attention to difficult-to-classify samples and optimize overall model performance. Experimental results show that compared to YOLOv8n, OEF-YOLO achieves improvements of 1.5 and 1.4 percentage points in terms of mAP@0.5 and mAP@0.5∶0.95, the parameters and computational complexity are 3.1×10⁶ and 6.5 GFLOPs. Frames Per Second (FPS) increases by 44 on a Graphic Processing Unit (GPU), achieving high-precision detection of fall events while also meeting deployment requirements in low computing scenarios.

Key words: object detection, lightweight, falling incidents, attention mechanism, Omni-dimensional Dynamic Convolution(ODConv)

宋杰, 徐慧英, 朱信忠, 黄晓, 陈晨, 王泽宇. 基于YOLOv8改进的跌倒检测算法: OEF-YOLO[J]. 计算机工程, 2025, 51(7): 127-139.

SONG Jie, XU Huiying, ZHU Xinzhong, HUANG Xiao, CHEN Chen, WANG Zeyu. Improved Fall Detection Algorithm Based on YOLOv8: OEF-YOLO[J]. Computer Engineering, 2025, 51(7): 127-139.

https://www.ecice06.com/CN/Y2025/V51/I7/127

图/表 16

图1 YOLOv8网络结构

Fig.1 Structure of YOLOv8 network

图2 OEF-YOLO结构

Fig.2 OEF-YOLO structure

图3 4种类型的关注标量作用示意图

Fig.3 Schematic diagram of the effects of four types of attention scalars

图4 EMA模块结构

Fig.4 Structure of EMA module

图5 C3模块和C2f模块结构

Fig.5 Structure of C3 module and C2f module

图6 C2f-ODConv模块结构

Fig.6 Structure of C2f-ODConv module

图7 OEF-YOLO颈部网络部分结构

Fig.7 Partial structure of OEF-YOLO neck network

图8 不同模型的检测精度和检测速度对比

Fig.8 Comparison of detection accuracy and detection speed of different models

图9 OEF-YOLO检测结果

Fig.9 Detection results of OEF-YOLO

图10 OEF-YOLO与YOLOv8n检测效果对比

Fig.10 Comparison of detection effects between OEF-YOLO and YOLOv8n

参考文献 41

1	闫玉娟, 李化, 赵菊敏, 等. 基于CRFID和模式识别的跌倒检测系统. 计算机工程, 2019, 45 (6): 297-302, 309. URL
	YAN Y J, LI H, ZHAO J M, et al. Fall detection system based on CRFID and pattern recognition. Computer Engineering, 2019, 45 (6): 297-302, 309. URL
2	陈文轩, 曾碧, 郭植星. 融合多特征与语义图卷积网络的摔倒检测方法. 计算机工程, 2023, 49 (5): 277-285, 294. URL
	CHEN W X, ZENG B, GUO Z X. Fall detection method integrating multi-feature and semantic graph convolution network. Computer Engineering, 2023, 49 (5): 277-285, 294. URL
3	陈明祥, 王钰, 刘环宇, 等. 基于人体稳定性的实时跌倒检测系统. 传感器与微系统, 2023, 42 (6): 17- 20. doi: 10.13873/J.1000-9787(2023)06-0017-04
	CHEN M X, YU Y, LIU H Y, et al. Real-time fall detection system based on human stability. Transducer and Microsystem Technologies, 2023, 42 (6): 17- 20. doi: 10.13873/J.1000-9787(2023)06-0017-04
4	RAZA A, YOUSAF M H, VELASTIN S A. Human fall detection using YOLO: a real-time and AI-on-the-edge perspective[C]//Proceedings of the 12th International Conference on Pattern Recognition Systems (ICPRS). Washington D. C., USA: IEEE Press, 2022: 1-6. URL
5	LONG K Z, HARON H, IBRAHIM M, et al. An image-based fall detection system using You Only Look Once (YOLO) algorithm to monitor elders' fall events[C]//Proceedings of Knowledge Management International Conference (KMICe). Berlin, Germany: Springer, 2021: 1-10. URL
6	YIN Y, LEI L, LIANG M, et al. Research on fall detection algorithm for the elderly living alone based on YOLO[C]//Proceedings of 2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT). Washington D. C., USA: IEEE Press, 2021: 403-408.
7	WANG X, JIA K. Human fall detection algorithm based on YOLOv3[C]//Proceedings of the 5th International Conference on Image, Vision and Computing. Washington D. C., USA: IEEE Press, 2020: 50-54. URL
8	LÜ X J, GAO Z L, YUAN C S, et al. Hybrid real-time fall detection system based on deep learning and multi-sensor fusion[C]//Proceedings of the 6th International Conference on Big Data and Information Analytics. Washington D. C., USA: IEEE Press, 2020: 386-391. URL
9	CHEN Y S, DU R X, LUO K T, et al. Fall detection system based on real-time pose estimation and SVM[C]//Proceedings of the 2nd International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE). Washington D. C., USA: IEEE Press, 2021: 990-993. URL
10	ZHENG H, LIU Y. Lightweight fall detection algorithm based on alphapose optimization model and ST-GCN. Mathematical Problems in Engineering, 2022, 2022 (1): 9962666.
11	REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39 (6): 1137- 1149. doi: 10.1109/TPAMI.2016.2577031
12	CAI Z W, VASCONCELOS N. Cascade R-CNN: delving into high quality object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2018: 6154-6162. URL
13	LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[EB/OL]. [2023-12-15]. https://link.springer.com/content/pdf/10.1007/978-3-319-46448-0_2.pdf. URL
14	赵冬冬, 谢墩翰, 陈朋, 等. 基于ZYNQ的轻量化YOLOv5声呐图像目标检测算法及实现. 光电工程, 2024, 51 (1): 230284. doi: 10.12086/oee.2024.230284
	ZHAO D D, XIE D H, CHEN P, et al. Lightweight YOLOv5 sonar image object detection algorithm and implementation based on ZYNQ. Opto-Electronic Engineering, 2024, 51 (1): 230284. doi: 10.12086/oee.2024.230284
15	LI C, ZHOU A J, YAO A H. Omni-dimensional dynamic convolution[EB/OL]. [2023-12-15]. https://arxiv.org/abs/2209.07947.
16	OUYANG D L, HE S, ZHANG G Z, et al. Efficient multi-scale attention module with cross-spatial learning[EB/OL]. [2023-12-15]. https://arxiv.org/abs/2305.13563?context=cs.AI. URL
17	ZHENG Z H, WANG P, REN D W, et al. Enhancing geometric factors in model learning and inference for object detection and instance segmentation. IEEE Transactions on Cybernetics, 2021, 52 (8): 8574- 8586. doi: 10.48550/arXiv.2005.03572
18	ZHANG Y F, REN W Q, ZHANG Z, et al. Focal and efficient IOU loss for accurate bounding box regression[EB/OL]. [2023-12-15]. https://arxiv.org/abs/2101.08158?context=cs.CV. URL
19	WANG C Y, BOCHKOVSKIY A, LIAO H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2023: 7464-7475. URL
20	LI C Y, LI L L, JIANG H L, et al. YOLOv6: a single-stage object detection framework for industrial applications[EB/OL]. [2023-12-15]. https://arxiv.org/abs/2209.02976.
21	WANG C Y, LIAO H Y M, WU Y H, et al. CSPNet: a new backbone that can enhance learning capability of CNN[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Washington D. C., USA: IEEE Press, 2020: 390-391. URL
22	LIU S, QI L, QIN H F, et al. Path aggregation network for instance segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2018: 8759-8768. URL
23	LIN T Y, DOLLÁR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2017: 2117-2125. URL
24	CHEN Y P, DAI X Y, LIU M C, et al. Dynamic convolution: attention over convolution kernels[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2020: 11030-11039. URL
25	HOU Q B, ZHOU D Q, FENG J S. Coordinate attention for efficient mobile network design[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2021: 13713-13722. URL
26	ZHANG X D, ZENG H, GUO S, et al. Efficient long-range attention network for image super-resolution[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2022: 649-667. URL
27	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2016: 770-778. URL
28	LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]//Proceedings of the IEEE International Conference on Computer Vision. Washington D. C., USA: IEEE Press, 2017: 2980-2988. URL
29	KWOLEK B, KEPSKI M. Human fall detection on embedded platform using depth maps and wireless accelerometer. Computer Methods and Programs in Biomedicine, 2014, 117 (3): 489- 501. doi: 10.1016/j.cmpb.2014.09.005
30	ADHIKARI K, BOUCHACHIA H, NAIT-CHARIF H. Activity recognition for indoor fall detection using convolutional neural network[C]//Proceedings of the 15th IAPR International Conference on Machine Vision Applications (MVA). Washington D. C., USA: IEEE Press, 2017: 81-84. URL
31	AUVINET E, ROUGIER C. Multiple cameras fall dataset[EB/OL]. [2023-12-15]. http://www.iro.umontreal.ca/~labimage/Dataset/.
32	CHEN J R, KAO S H, HE H, et al. Run, don't walk: chasing higher FLOPS for faster neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2023: 12021-12031. URL
33	DING X H, ZHANG X Y, HAN J G, et al. Diverse branch block: building a convolution as an inception-like unit[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2021: 10886-10895. URL
34	ZHENG Z H, WANG P, LIU W, et al. Distance-IoU loss: faster and better learning for bounding box regression[EB/OL]. [2023-12-15]. https://arxiv.org/abs/1911.08287.
35	REZATOFIGHI H, TSOI N, GWAK J Y, et al. Generalized intersection over union: a metric and a loss for bounding box regression[EB/OL]. [2023-12-15]. https://arxiv.org/abs/1902.09630. URL
36	GEVORGYAN Z. SIoU loss: more powerful learning for bounding box regression[EB/OL]. [2023-12-15]. https://arxiv.org/abs/2205.12740.
37	ZHAO Y, LÜ W, XU S, et al. DETRs beat YOLOs on real-time object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2024: 16965-16974.
38	ADARSH P, RATHI P, KUMAR M. YOLOv3-Tiny: object detection and recognition using one stage improved model[C]//Proceedings of the 6th International Conference on Advanced Computing and Communication Systems (ICACCS). Washington D. C., USA: IEEE Press, 2020: 687-694. URL
39	BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLOv4: optimal speed and accuracy of object detection[EB/OL].[2023-12-15]. https://arxiv.org/abs/2004.10934. URL
40	WANG C Y, YEH I H, LIAO H Y M. YOLOv9: learning what you want to learn using programmable gradient information[EB/OL].[2023-12-15]. https://link.springer.com/chapter/10.1007/978-3-031-72751-1_1.
41	朱胜豪, 钱承山, 阚希. 改进YOLOv5的高精度跌倒检测算法. 计算机工程与应用, 2024, 60 (11): 105- 114. doi: 10.3778/j.issn.1002-8331.2307-0190
	ZHU S H, QIAN C S, KAN X. High-precision fall detection algorithm with improved YOLOv5. Computer Engineering and Application, 2024, 60 (11): 105- 114. doi: 10.3778/j.issn.1002-8331.2307-0190

[1]	刘春霞, 孟吉星, 潘理虎, 龚大立. 融合RGB与IR图像的遥感小目标检测方法[J]. 计算机工程, 2025, 51(7): 326-338.
[2]	张佳承, 韦锦, 陈义时. 改进YOLOv8的实时轻量化鲁棒绿篱检测算法[J]. 计算机工程, 2025, 51(7): 362-374.
[3]	栾孟娜, 郑秋梅, 王风华. 基于DMC-YOLO的交通标志实时检测算法[J]. 计算机工程, 2025, 51(7): 90-99.
[4]	彭菊红, 张弛, 高谦, 张光明, 谈栋华, 赵明俊. 基于改进的YOLOv8算法的钢材缺陷检测[J]. 计算机工程, 2025, 51(7): 152-160.
[5]	刘旭东, 杨绪兵. L1-OCSVM模型设计及其在林业目标检测中的应用[J]. 计算机工程, 2025, 51(7): 375-384.
[6]	单鹏畅, 高利剑, 董文龙, 毛启容. 基于显著目标追踪的行为检测方法[J]. 计算机工程, 2025, 51(6): 93-101.
[7]	华家宝, 张京瑞, 朱福民, 陈璐. 基于路侧相机的自适应空间变换车辆检测方法[J]. 计算机工程, 2025, 51(6): 349-359.
[8]	冯晓飞, 谢诚, 张秀振, 董仕奎, 陈军胜, 叶舒, 钟忺. 基于动静结合互学习的预制梁工序检测方法[J]. 计算机工程, 2025, 51(6): 385-394.
[9]	刘凯, 任洪逸, 李蓥, 季怡, 刘纯平. 基于交叉模态注意力特征增强的医学视觉问答[J]. 计算机工程, 2025, 51(6): 49-56.
[10]	奚琦, 王明杰, 魏敬和, 赵伟. 基于改进YOLOv3的航拍小目标检测算法[J]. 计算机工程, 2025, 51(6): 184-192.
[11]	赵小虎, 谢礼逊, 慕灯聪, 张悦. 基于TCM-YOLO网络的金属表面缺陷检测方法[J]. 计算机工程, 2025, 51(6): 338-348.
[12]	李毅, 徐慧英, 朱信忠, 黄晓, 王舒梦, 李悉钰. 基于YOLOv5n模型改进的口罩检测算法: Mask-YOLO[J]. 计算机工程, 2025, 51(6): 297-310.
[13]	黄琦强, 安国成, 熊刚. 基于视觉-语言预训练模型的开集交通目标检测算法[J]. 计算机工程, 2025, 51(6): 375-384.
[14]	王晓龙, 江波, 罗润书, 安国成. 基于多信息融合的高速收费站拥堵检测算法[J]. 计算机工程, 2025, 51(5): 377-386.
[15]	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.

选择文件类型/文献管理软件名称

选择包含的内容