Classification of sows' activity types from acceleration patterns using univariate and multivariate models

Automatic monitoring of animal behavior in livestock production opens up possibilities for on-line monitoring of, among others, oestrus, health disorders, and animal welfare in general. The aim of this study is to use time series of acceleration measurements in order to automatically classify activity types performed by group-housed sows. Extracts of series collected for 11 sows are associated with 5 activity types: feeding (FE), rooting (RO), walking (WA), lying sternally (LS) and lying laterally (LL). A total of 24h of three-dimensional series is used. One univariate model and four multivariate models are used to describe all five activity types. Three multivariate models differ in their variance/covariance structure; a fourth alternative multivariate model (MU) simply combines the 3-axes of the univariate model, assuming independence. For each model, the activity-specific parameters are estimated using the EM algorithm. The classification method, based on a Multi-Process Kalman Filter provides posterior probabilities for each of the 5 activities, for a given series. For the univariate model, LL is the activity which is best recognized by the 3-axes; FE, RO and WA are best recognized by one particular axis; LS is poorest recognized. The average results are improved by using all four types of multivariate models. The percentages of activity recognition are similar among the multivariate models. By grouping the activity types into active (FE, RO, WA) vs. passive (LS, LL) categories, the method allows to correctly classify 96% of the active category and 94% of the passive category.

[1]  Cécile Cornou,et al.  Classifying sows' activity types from acceleration patterns An application of the Multi-Process Kalman Filter , 2008 .

[2]  Ulrich Brehme,et al.  ALT pedometer-New sensor-aided measurement system for improvement in oestrus detection , 2008 .

[3]  Anders Ringgaard Kristensen,et al.  Modelling the drinking patterns of young pigs using a state space model , 2005 .

[4]  B. Jørgensen,et al.  State‐space models for multivariate longitudinal data of mixed types , 1996 .

[5]  T. Leroy,et al.  Automatic real-time monitoring of locomotion and posture behaviour of pregnant cows prior to calving using online image analysis , 2008 .

[6]  Thomas Bak,et al.  Monitoring cow behavior parameters based on received signal strenght using wireless sensor networks , 2007 .

[7]  M. West,et al.  Bayesian forecasting and dynamic models , 1989 .

[8]  J. P. Holland,et al.  An automated sensor-based method of simple behavioural classification of sheep in extensive systems , 2008 .

[9]  Iver Thysen,et al.  Monitoring Bulk Tank Somatic Cell Counts by a Multi-Process Kalman Filter , 1993 .

[10]  B Erez,et al.  A microcomputer-photocell system to monitor periparturient activity of sows and transfer data to remote location. , 1990, Journal of animal science.

[11]  Cécile Cornou,et al.  Automated oestrus detection methods in group housed sows: Review of the current methods and perspectives for development , 2006 .

[12]  A. Feinstein.  XXXIV. The other side of ‘statistical significance’: alpha, beta. delta, and the calculation of sample size , 1975, Clinical pharmacology and therapeutics.

[13]  Sidney Cox,et al.  Precision livestock farming. , 2003 .

[14]  R. Firk,et al.  Automation of oestrus detection in dairy cows: a review , 2002 .

[15]  Michael A. West,et al.  Bayesian Forecasting and Dynamic Models (2nd edn) , 1997, J. Oper. Res. Soc..

[16]  Jukka Heikkonen,et al.  Using movement sensors to detect the onset of farrowing , 2008 .