首页 | 官方网站   微博 | 高级检索  
     

可穿戴生物感知的肉羊运输应激特征参数提取与建模
引用本文:马瑞芹,张梦杰,于瑞航,崔衍,王想.可穿戴生物感知的肉羊运输应激特征参数提取与建模[J].农业机械学报,2021,52(10):291-299,416.
作者姓名:马瑞芹  张梦杰  于瑞航  崔衍  王想
作者单位:中国农业大学
基金项目:国家重点研发计划——政府间国际合作项目(2017YFE0111200)
摘    要:传统的肉羊应激反应检测方法多采用静态采血化验法,在动物养殖运输过程中无法做到实时且连续地检测,为此设计了适用于肉羊运输过程中的可穿戴生物与环境信号的动态检测装置,利用脉搏波传感器和红外温度传感器获取肉羊运输过程中的心率和体温,采用时域分析方法提取肉羊运输应激特征参数中5个典型特征参数,并对其进行特征分析。特征参数MR、SDNN、RMSSD、pNN50、CV随运输时间的延长呈增加趋势,并且由于道路的变化而具有随机性,证明了运输应激的变化;基于专家经验与有监督的机器学习方法构建肉羊运输过程应激状态预测模型,模型的平均预测准确率在89.81%以上。结果表明,本文设计的可穿戴生物感知装置,能够有效地识别出运输应激状态,实现肉羊运输应激分级预测管理机制,解决了肉羊运输监管与品控等问题。

关 键 词:可穿戴  生物感知  肉羊运输  应激检测  特征提取
收稿时间:2021/6/25 0:00:00

Feature Parameter Extraction and Modeling of Stress Response Based on Wearable Biosensing for Mutton Sheep Transportation
MA Ruiqin,ZHANG Mengjie,YU Ruihang,CUI Yan,WANG Xiang.Feature Parameter Extraction and Modeling of Stress Response Based on Wearable Biosensing for Mutton Sheep Transportation[J].Transactions of the Chinese Society of Agricultural Machinery,2021,52(10):291-299,416.
Authors:MA Ruiqin  ZHANG Mengjie  YU Ruihang  CUI Yan  WANG Xiang
Affiliation:China Agricultural University
Abstract:The traditional detection methods of stress response of mutton sheep focus on the static blood test method, which cannot achieve real-time and continuous detection in the process of animal farming and transportation. A set of devices for dynamic detection of wearable biological and environmental signals in the process of mutton sheep transportation was designed. The heart rate and body temperature of mutton sheep during transportation were obtained by pulse wave sensor and infrared temperature sensor. The time domain analysis method was used to extract five typical characteristic parameters of transport stress characteristic parameters of mutton sheep, and their characteristics were analyzed in detail. The characteristic parameters MR, SDNN, RMSSD, pNN50 and CV tended to increase with the extension of transportation time, and were random due to changes in roads, which proved the changes in transportation stress. The average prediction accuracy of the stress state prediction model based on expert experience and supervised machine learning method was more than 89.81%. The results showed that through the development of wearable biosensing devices the pulse wave and body temperature can be monitored and the characteristic parameters of transport stress of mutton sheep can be extracted. The three states of transport stress can be effectively identified, the classification prediction management mechanism of transport stress of mutton sheep can be realized, and the problems such as low efficiency of transport supervision and quality control of mutton sheep can be solved.
Keywords:wearable  biosensing  mutton sheep transportation  stress detection  feature extraction
点击此处可从《农业机械学报》浏览原始摘要信息
点击此处可从《农业机械学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号