Improved pig behavior analysis by optimizing window sizes for individual behaviors on acceleration and angular velocity data.
Improved pig behavior analysis by optimizing window sizes for individual behaviors on acceleration and angular velocity data.
复制标题
DOI：
10.1093/jas/skac293
复制
发表时间：
2022-11-01
影响因子：
3.3
通讯作者：
Ha, Sook S.
中科院分区：
文献类型：
--
作者：
Alghamdi, Saleh;Zhao, Zhuqing;Ha, Dong S.;Morota, Gota;Ha, Sook S.
关键词：
本文介绍了机器学习算法的应用，从安装在猪身上的无线传感器节点收集的数据中识别猪的行为。连接在猪背上的传感器节点在三个轴上感知加速度和角速度，并将感知到的数据无线传输到主机上。两个摄像机，一个安装在猪圈的天花板上，另一个安装在围栏上，为数据注释提供了地面实况。由于典型的行为周期取决于行为类型，我们对不同窗口大小（WS）和步长（SS）的加速数据进行了分割，并测试了不同WS和SS对不同活动的分类性能的影响。在探索可能的组合后，我们选择了最佳WS和SS。为了比较性能，我们使用了五种机器学习算法，特别是支持向量机，k近邻，决策树，朴素贝叶斯和随机森林（RF）。在5种算法中，RF在4个主要行为F1得分最高，占92.36%。该算法的F1得分为“吃”为0.98，“躺”为0.99，“走”为0.93，“站”为0.91。“进食”和“躺着”的最佳WS为7秒，“行走”和“站立”的最佳WS为3秒。提出的工作表明，基于行为的长度，自适应窗口和步长提高了分类性能。我们的贡献是使用自适应窗口和步长来对不同持续时间的行为进行分类，以便窗口可以精确地捕获各种行为的持续时间，并增加段的数量以减轻类分布的倾斜。研究表明，基于行为长度的最佳窗口和步长提高了分类性能。
This paper presents the application of machine learning algorithms to identify pigs’ behaviors from data collected using the wireless sensor nodes mounted on pigs. The sensor node attached to a pig’s back senses the acceleration and angular velocity in three axes, and the sensed data are transmitted to a host computer wirelessly. Two video cameras, one attached to the ceiling of the pigpen and the other one to a fence, provided ground truth for data annotations. The data were collected from pigs for 131 h over 2 mo. As the typical behavior period depends on the behavior type, we segmented the acceleration data with different window sizes (WS) and step sizes (SS), and tested how the classification performance of different activities varied with different WS and SS. After exploring the possible combinations, we selected the optimum WS and SS. To compare performance, we used five machine learning algorithms, specifically support vector machine, k-nearest neighbors, decision trees, naive Bayes, and random forest (RF). Among the five algorithms, RF achieved the highest F1 score for four major behaviors consisting of 92.36% in total. The F1 scores of the algorithm were 0.98 for “eating,” 0.99 for “lying,” 0.93 for “walking,” and 0.91 for “standing” behaviors. The optimal WS was 7 s for “eating” and “lying,” and 3 s for “walking” and “standing.” The proposed work demonstrates that, based on the length of behavior, the adaptive window and step sizes increase the classification performance. Our contribution is using the adaptive window and step sizes to classify behaviors with different durations so that the windows could precisely capture the duration of various behaviors and increase the number of segments to alleviate the skewed class distribution. The proposed work demonstrates that the optimal window and step sizes based on the length of behavior increased the classification performances.
登录
查看更多内容
DOI：
10.1016/b978-0-12-818366-3.00005-8
发表时间：
2020-01-01
期刊：
DATA DEMOCRACY: AT THE NEXUS OF ARTIFICIAL INTELLIGENCE, SOFTWARE DEVELOPMENT, AND KNOWLEDGE ENGINEERING
影响因子：
--
作者：
Kulkarni, Ajay;Chong, Deri;Batarseh, Feras A.
通讯作者：
Batarseh, Feras A.
DOI：
10.3390/s140406474
发表时间：
2014-04-09
期刊：
Sensors (Basel, Switzerland)
影响因子：
--
作者：
Banos O;Galvez JM;Damas M;Pomares H;Rojas I
通讯作者：
Rojas I
DOI：
10.3390/s18092946
发表时间：
2018-09-04
期刊：
Sensors (Basel, Switzerland)
影响因子：
--
作者：
Syafrudin M;Alfian G;Fitriyani NL;Rhee J
通讯作者：
Rhee J
影响因子：
2.3
作者：
Cornou, Cecile;Lundbye-Christensen, Soren
通讯作者：
Lundbye-Christensen, Soren
影响因子：
3.9
作者：
Huang, Jianglin;Li, Yan-Fu;Xie, Min
通讯作者：
Xie, Min