母羊产前行为特征分析与识别—基于可穿戴检测装置构架

发布时间：2018-04-27 20:00

本文选题：母羊产前行为 + 加速度传感器　；参考：《农机化研究》2017年09期

【摘要】：准确获取母羊临产前行为活动方式及其规律对预测母羊分娩时间、判断其在分娩时是否需要提供人工助产具有重要的意义。目前,在大规模集约化养殖环境下大多依靠饲养员对母羊产前行为进行连续观察,耗时耗力,且主观性较大。为此,设计了一种基于可穿戴数据检测装置的母羊产前行为特征实时监测系统。该装置以单片机STM8为主控芯片,采用MPU6050三轴加速度传感器获取母羊产前行为数据,通过Wifi无线传输模块USR-C2 1 5将数据传输至上位机。利用K均值聚类算法对母羊产前3种行为(趴卧、行走、站立)进行特征分析与识别。试验表明:系统可以快速、准确的获取母羊产前行为的加速度信息,并对母羊的3种产前行为能够正确的分类识别,识别率为82.69%,能够满足母羊产前行为识别的要求。
[Abstract]:It is of great significance to obtain accurately the behavior pattern and its regularity of ewe before delivery to predict the delivery time of ewe and to judge whether it is necessary to provide artificial midwifery during delivery. At present, in the large-scale intensive farming environment, most of them rely on the breeder to observe the antenatal behavior of ewe continuously, which is time-consuming and labor-consuming, and subjective. Therefore, a real-time monitoring system based on wearable data detection device for ewe antenatal behavior is designed. The device uses single chip computer (STM8) as main control chip, adopts MPU6050 three-axis acceleration sensor to acquire ewe prenatal behavior data, and transmits the data to upper computer through Wifi wireless transmission module USR-C2 15. K-means clustering algorithm was used to analyze and identify the characteristics of three prenatal behaviors (lying down, walking and standing) of ewe. The test results show that the system can obtain the acceleration information of ewe antenatal behavior quickly and accurately, and can correctly classify and recognize the three antenatal behaviors of ewe, and the recognition rate is 82.69, which can meet the requirements of ewe prenatal behavior recognition.
【作者单位】：内蒙古农业大学机电工程学院;内蒙古师范大学物理与电子信息学院;
【基金】：国家自然科学基金项目(61461042;11364029) 内蒙古自然科学基金项目(2012MS0720) 内蒙古“草原英才”产业创新人才团队项目(内组通字[2014]27号)
【分类号】：TP274;S826

【相似文献】