电子测量与仪器学报

刘韵婷,刘欣然,肖培宇,李福望,王晓艺.基于改进YOLOv8室内老人跌倒检测算法研究[J].电子测量与仪器学报,2026,40(2):164-174

基于改进YOLOv8室内老人跌倒检测算法研究

Research on improved YOLOv8 algorithm for indoor elderly fall detection

DOI：

中文关键词: 跌倒检测 YOLOv8 ODConv FasterNet LSKA SEAM

英文关键词:fall detection YOLOv8 ODConv FasterNet LSKA SEAM

基金项目:国家重点研发计划子课题(2017YFC0821001-2)、国家重点研发计划子课题（2020YFC2006701-1）项目资助

作者	单位
刘韵婷	沈阳理工大学自动化与电气工程学院沈阳110159
刘欣然	沈阳理工大学自动化与电气工程学院沈阳110159
肖培宇	沈阳理工大学自动化与电气工程学院沈阳110159
李福望	沈阳理工大学自动化与电气工程学院沈阳110159
王晓艺	沈阳理工大学自动化与电气工程学院沈阳110159

Author	Institution
Liu Yunting	School of Automation and Electrical Engineering，Shenyang Ligong University, Shenyang 110159，China
Liu Xinran	School of Automation and Electrical Engineering，Shenyang Ligong University, Shenyang 110159，China
Xiao Peiyu	School of Automation and Electrical Engineering，Shenyang Ligong University, Shenyang 110159，China
Li Fuwang	School of Automation and Electrical Engineering，Shenyang Ligong University, Shenyang 110159，China
Wang Xiaoyi	School of Automation and Electrical Engineering，Shenyang Ligong University, Shenyang 110159，China

摘要点击次数: 440

全文下载次数: 221

中文摘要:

针对传统跌倒检测在遮挡及光照干扰下易误检、难以平衡轻量化和检测精度的问题，提出了一种基于改进YOLOv8的室内老人跌倒检测算法。首先，设计了将全维动态卷积融合骨干网络的C2f模块中以此在特征提取阶段具备自适应性，提升了模型的特征提取能力。在颈部网络采用FasterNet模块改进C2f模块，有效降低了计算成本；同时，在快速空间金字塔池化（SPPF）中融合大核可分离卷积注意力机制（LSKA），提高了目标检测的精确度，并在检测头前引入SEAM（spatial-enhanced attention module）注意力机制，进一步增强了检测能力。对模型进行了对比实验、消融实验，并利用改进后的算法将检测结果转化为热度图进一步验证算法的有效性。实验结果表明，与原始YOLOv8n模型相比，改进后的算法平均精度均值（mAP）mAP@0.5达到91.0%，提升了1.1%，参数量减少了20.6%，浮点计算量减少了42.7%，证明所提算法在室内老人跌倒检测任务中实现了检测精度与模型轻量化之间的有效平衡。

英文摘要:

To address the limitations of traditional fall detection methods, which are prone to false detections under occlusion and illumination interference and struggle to balance lightweight design with detection accuracy, this paper proposes an improved YOLOv8-based algorithm for indoor elderly fall detection. Specifically, omni-dimensional dynamic convolution is integrated into the C2f module of the backbone network, enabling adaptive feature extraction and enhancing representational capability. In the neck network, the C2f module is further optimized with the FasterNet module to effectively reduce computational cost. In addition, a large selective kernel attention (LSKA) mechanism is embedded into the SPPF module to improve detection precision, while a spatial-enhanced attention module (SEAM) is introduced into the detection head to further strengthen discriminative ability. Comparative and ablation experiments were conducted, and the detection results were further visualized as heatmaps to validate the effectiveness of the proposed approach. Experimental results demonstrate that, compared with the baseline YOLOv8n model, the improved algorithm achieves an mAP@0.5 of 91.0% (an improvement of 1.1%), with a 20.6% reduction in parameters and a 42.7% decrease in GFLOPs, thereby confirming that the proposed method effectively balances detection accuracy and lightweight design in indoor elderly fall detection tasks.

查看全文查看/发表评论下载PDF阅读器