• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

一种基于地板声学特征的用于用户辅助人体跌倒检测的单类支持向量机与模板匹配相结合的方法。

A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features.

作者信息

Droghini Diego, Ferretti Daniele, Principi Emanuele, Squartini Stefano, Piazza Francesco

机构信息

Department of Information Engineering, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, Italy.

出版信息

Comput Intell Neurosci. 2017;2017:1512670. doi: 10.1155/2017/1512670. Epub 2017 May 30.

DOI:10.1155/2017/1512670
PMID:28638405
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC5468803/
Abstract

The primary cause of injury-related death for the elders is represented by falls. The scientific community devoted them particular attention, since injuries can be limited by an early detection of the event. The solution proposed in this paper is based on a combined One-Class SVM (OCSVM) and template-matching classifier that discriminate human falls from nonfalls in a semisupervised framework. Acoustic signals are captured by means of a Floor Acoustic Sensor; then Mel-Frequency Cepstral Coefficients and Gaussian Mean Supervectors (GMSs) are extracted for the fall/nonfall discrimination. Here we propose a single-sensor two-stage user-aided approach: in the first stage, the OCSVM detects abnormal acoustic events. In the second, the template-matching classifier produces the final decision exploiting a set of template GMSs related to the events marked as false positives by the user. The performance of the algorithm has been evaluated on a corpus containing human falls and nonfall sounds. Compared to the OCSVM only approach, the proposed algorithm improves the performance by 10.14% in clean conditions and 4.84% in noisy conditions. Compared to Popescu and Mahnot (2009) the performance improvement is 19.96% in clean conditions and 8.08% in noisy conditions.

摘要

老年人与伤害相关的死亡主要原因是跌倒。科学界对其给予了特别关注，因为通过早期发现该事件可以减少伤害。本文提出的解决方案基于一种组合的单类支持向量机（OCSVM）和模板匹配分类器，该分类器在半监督框架下区分人类跌倒和未跌倒情况。声学信号通过地面声学传感器采集；然后提取梅尔频率倒谱系数和高斯均值超向量（GMSs）用于跌倒/未跌倒判别。在此，我们提出一种单传感器两阶段用户辅助方法：在第一阶段，OCSVM检测异常声学事件。在第二阶段，模板匹配分类器利用与用户标记为误报的事件相关的一组模板GMSs做出最终决策。该算法的性能已在一个包含人类跌倒声音和未跌倒声音的语料库上进行了评估。与仅使用OCSVM的方法相比，所提出的算法在干净条件下性能提高了10.14%，在有噪声条件下提高了4.84%。与Popescu和Mahnot（2009）的方法相比，在干净条件下性能提高了19.96%，在有噪声条件下提高了8.08%。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/14622d1505ca/CIN2017-1512670.010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/eef63592f82f/CIN2017-1512670.001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/d35a785599e0/CIN2017-1512670.002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/463dfa774d10/CIN2017-1512670.003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/b593262639a0/CIN2017-1512670.004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/cd3129e99b77/CIN2017-1512670.005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/c8a56566e2c2/CIN2017-1512670.006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/1b45d5da7ceb/CIN2017-1512670.007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/ad657d800315/CIN2017-1512670.008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/83d7d4bcbe0f/CIN2017-1512670.009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/14622d1505ca/CIN2017-1512670.010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/eef63592f82f/CIN2017-1512670.001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/d35a785599e0/CIN2017-1512670.002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/463dfa774d10/CIN2017-1512670.003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/b593262639a0/CIN2017-1512670.004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/cd3129e99b77/CIN2017-1512670.005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/c8a56566e2c2/CIN2017-1512670.006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/1b45d5da7ceb/CIN2017-1512670.007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/ad657d800315/CIN2017-1512670.008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/83d7d4bcbe0f/CIN2017-1512670.009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/67f3/5468803/14622d1505ca/CIN2017-1512670.010.jpg

相似文献

1
A Combined One-Class SVM and Template-Matching Approach for User-Aided Human Fall Detection by Means of Floor Acoustic Features.一种基于地板声学特征的用于用户辅助人体跌倒检测的单类支持向量机与模板匹配相结合的方法。
Comput Intell Neurosci. 2017;2017:1512670. doi: 10.1155/2017/1512670. Epub 2017 May 30.
2
A microphone array system for automatic fall detection.用于自动跌倒检测的麦克风阵列系统。
IEEE Trans Biomed Eng. 2012 May;59(5):1291-301. doi: 10.1109/TBME.2012.2186449.
3
Acoustic fall detection using one-class classifiers.使用单类分类器的声学跌倒检测。
Annu Int Conf IEEE Eng Med Biol Soc. 2009;2009:3505-8. doi: 10.1109/IEMBS.2009.5334521.
4
A comparative study of the SVM and K-nn machine learning algorithms for the diagnosis of respiratory pathologies using pulmonary acoustic signals.基于肺部声信号的呼吸病理诊断的支持向量机（SVM）和 K-近邻（K-nn）机器学习算法的比较研究。
BMC Bioinformatics. 2014 Jun 27;15:223. doi: 10.1186/1471-2105-15-223.
5
Validation of accuracy of SVM-based fall detection system using real-world fall and non-fall datasets.使用真实世界的跌倒和非跌倒数据集验证基于支持向量机的跌倒检测系统的准确性。
PLoS One. 2017 Jul 5;12(7):e0180318. doi: 10.1371/journal.pone.0180318. eCollection 2017.
6
Fall Detection Using Smartphone Audio Features.利用智能手机音频特征进行跌倒检测。
IEEE J Biomed Health Inform. 2016 Jul;20(4):1073-80. doi: 10.1109/JBHI.2015.2425932. Epub 2015 Apr 23.
7
Improving automatic sound-based fall detection using iVAT clustering and GA-based feature selection.使用iVAT聚类和基于遗传算法的特征选择改进基于声音的自动跌倒检测。
Annu Int Conf IEEE Eng Med Biol Soc. 2012;2012:5867-70. doi: 10.1109/EMBC.2012.6347328.
8
Accelerometer and Camera-Based Strategy for Improved Human Fall Detection.基于加速度计和摄像头的改进型人体跌倒检测策略。
J Med Syst. 2016 Dec;40(12):284. doi: 10.1007/s10916-016-0639-6. Epub 2016 Oct 29.
9
An online one class support vector machine-based person-specific fall detection system for monitoring an elderly individual in a room environment.一种基于在线单类支持向量机的个体特定跌倒检测系统，用于在房间环境中监测老年人。
IEEE J Biomed Health Inform. 2013 Nov;17(6):1002-14. doi: 10.1109/JBHI.2013.2274479.
10
Detection of Human Fall Using Floor Vibration and Multi-Features Semi-Supervised SVM.基于地板振动和多特征半监督 SVM 的人体跌倒检测。
Sensors (Basel). 2019 Aug 28;19(17):3720. doi: 10.3390/s19173720.

引用本文的文献

1
System-level interventions for fall risk assessment, fall prevention and fall injury prevention in long-term care facilities: a scoping review.长期护理机构中跌倒风险评估、跌倒预防和跌倒伤害预防的系统层面干预措施：一项范围综述
Inj Prev. 2024 Oct 30. doi: 10.1136/ip-2024-045386.
2
Markerless Radio Frequency Indoor Monitoring for Telemedicine: Gait Analysis, Indoor Positioning, Fall Detection, Tremor Analysis, Vital Signs and Sleep Monitoring.无标记射频室内远程医疗监测：步态分析、室内定位、跌倒检测、震颤分析、生命体征和睡眠监测。
Sensors (Basel). 2022 Nov 4;22(21):8486. doi: 10.3390/s22218486.
3
IR-UWB Sensor Based Fall Detection Method Using CNN Algorithm.

本文引用的文献

1
A wavelet-based approach to fall detection.基于小波的跌倒检测方法。
Sensors (Basel). 2015 May 20;15(5):11575-86. doi: 10.3390/s150511575.
2
Fall Detection Using Smartphone Audio Features.利用智能手机音频特征进行跌倒检测。
IEEE J Biomed Health Inform. 2016 Jul;20(4):1073-80. doi: 10.1109/JBHI.2015.2425932. Epub 2015 Apr 23.
3
Automatic fall monitoring: a review.自动跌倒监测：综述
基于卷积神经网络算法的 IR-UWB 传感器跌倒检测方法。
Sensors (Basel). 2020 Oct 21;20(20):5948. doi: 10.3390/s20205948.
4
AI-Based Early Change Detection in Smart Living Environments.智能生活环境中基于人工智能的早期变化检测
Sensors (Basel). 2019 Aug 14;19(16):3549. doi: 10.3390/s19163549.
5
A Radar-Based Smart Sensor for Unobtrusive Elderly Monitoring in Ambient Assisted Living Applications.基于雷达的智能传感器，用于在安闲辅助生活应用中进行非干扰性老年人监测。
Biosensors (Basel). 2017 Nov 24;7(4):55. doi: 10.3390/bios7040055.
Sensors (Basel). 2014 Jul 18;14(7):12900-36. doi: 10.3390/s140712900.
4
Efficient source separation algorithms for acoustic fall detection using a microsoft kinect.利用微软 Kinect 实现声学跌倒检测的高效声源分离算法。
IEEE Trans Biomed Eng. 2014 Mar;61(3):745-55. doi: 10.1109/TBME.2013.2288783. Epub 2013 Nov 5.
5
A microphone array system for automatic fall detection.用于自动跌倒检测的麦克风阵列系统。
IEEE Trans Biomed Eng. 2012 May;59(5):1291-301. doi: 10.1109/TBME.2012.2186449.
6
A review of accelerometry-based wearable motion detectors for physical activity monitoring.基于加速计的可穿戴运动探测器在身体活动监测中的应用综述。
Sensors (Basel). 2010;10(8):7772-88. doi: 10.3390/s100807772. Epub 2010 Aug 20.
7
Emergency fall incidents detection in assisted living environments utilizing motion, sound, and visual perceptual components.利用运动、声音和视觉感知组件在辅助生活环境中进行紧急跌倒事件检测。
IEEE Trans Inf Technol Biomed. 2011 Mar;15(2):277-89. doi: 10.1109/TITB.2010.2091140. Epub 2010 Nov 9.
8
Acoustic fall detection using one-class classifiers.使用单类分类器的声学跌倒检测。
Annu Int Conf IEEE Eng Med Biol Soc. 2009;2009:3505-8. doi: 10.1109/IEMBS.2009.5334521.
9
A method for automatic fall detection of elderly people using floor vibrations and sound--proof of concept on human mimicking doll falls.利用地板振动和声音进行老年人自动摔倒检测的方法——在模拟人摔倒方面的概念验证。
IEEE Trans Biomed Eng. 2009 Dec;56(12):2858-67. doi: 10.1109/TBME.2009.2030171. Epub 2009 Aug 25.
10
Fall detection--principles and methods.跌倒检测——原理与方法
Annu Int Conf IEEE Eng Med Biol Soc. 2007;2007:1663-6. doi: 10.1109/IEMBS.2007.4352627.