Get access to all handy features included in the IVIS website
- Get unlimited access to books, proceedings and journals.
- Get access to a global catalogue of meetings, on-site and online courses, webinars and educational videos.
- Bookmark your favorite articles in My Library for future reading.
- Save future meetings and courses in My Calendar and My e-Learning.
- Ask authors questions and read what others have to say.
Use of a real-time location system to detect cows in distinct functional areas within a barn
Chapa, J.M.; Lidauer, L.; Berger, A...
Get access to all handy features included in the IVIS website
- Get unlimited access to books, proceedings and journals.
- Get access to a global catalogue of meetings, on-site and online courses, webinars and educational videos.
- Bookmark your favorite articles in My Library for future reading.
- Save future meetings and courses in My Calendar and My e-Learning.
- Ask authors questions and read what others have to say.
Read
Objectives: Automated monitoring of animals by use of various sensor technologies is already used for many decades on dairy farms. Recent research and development of new sensor technologies and features aim to further improve animal health, welfare, and management procedures. Modern sensor technologies allow, among others, the tracking of animals in real-time in a barn. This can be used, for example, to estimate the time an animal spend at relevant ‘functional areas’ such as the feed bunk, cubicles, or alleys. The resulting data can then potentially be used as early indicators for disease, discomfort and to estimate the welfare status of an animal.
In this field study, we tested the real-time localization sys- tem (RTLS) of a commercially available system (SMARTBOW, Smartbow/Zoetis LLC, Weibern, Austria) to detect animals in predefined functional areas. The system consists of an ear-attached accelerometer that sends low-frequency signals to receivers, which transmit the data to a local farm server. Based on the incoming data of individual animals, the server software triangulates the location of an animal within the barn in real-time. The objective of this study was to determine the accuracy of the system to predict the location of the cow and the agreement between visual observations (VO) and RTLS observations for the total time spent by cows in relevant areas of the barn.
Material and methods: The study was conducted in May 2019 on a commercial dairy farm in Austria, housing approximately 35 Brown Swiss cows. The SMARTBOW (SB) tags were attached to the left ears of the animals. In advance of the study, functional areas of interest (i.e. the feed bunk, cubicles, and alleys) were predefined in the software of the sensor system. Cows were video recorded for three consecutive days using 9 digital cameras (DS-2CD2642FWD-IZS, Hikvision, Hangzhou, China). From these recordings, approximately 1 h was selected randomly each day for every cow (3 d × 35 cows). For each minute of an hour, animal position within a specific functional area was visually observed and labeled by use of specialized software for video analyses (Mangold Interact, Mangold International GmbH, Arnstorf, Germany). Data of the video observations (VO) served as gold standard in this study. A total of 6,030 pairs of location data, derived from VO and the SB system, were used for statistical analyses. Categorical data were used to estimate the agreement between the two methods. For each functional area of interest, the sensitivity (Se), specificity (Sp), and accuracy (Acc) were calculated. The total time spent (min/h) per cow in the specific areas was analyzed using Spearman correlations.
Results: Overall, a Cohen ́s kappa of 0.78, indicating a ‘substantial agreement’, between VO and SB was obtained. Se and Sp were determined for locating the cows in the alley (74.0 and 91.2%), feed bunk (93.5 and 86.2%), and cubicle (90.5 and 83.3%), respectively, and overall accuracy of 87.6%. The correlation between VO and SB for the ‘total time’ an animal spent within an hour in alleys, at the feed bunk, and in cubicles was ‘good’ to ‘strong’ with correlation coefficients of 0.82, 0.98, and 0.92, respectively.
Conclusions: Overall, the real-time localization feature from SB was successful in predicting the position of an animal in a specific functional area. The estimated times, which an animal spent per hour in these specific areas were good. Future research should focus on, whether these times could be used as early indicators for disease, discomfort and to estimate the welfare status of an animal.
Keywords: Animal tracking, dairy cow, precision dairy farming, real-time location system, time budget.
Get access to all handy features included in the IVIS website
- Get unlimited access to books, proceedings and journals.
- Get access to a global catalogue of meetings, on-site and online courses, webinars and educational videos.
- Bookmark your favorite articles in My Library for future reading.
- Save future meetings and courses in My Calendar and My e-Learning.
- Ask authors questions and read what others have to say.
About
Affiliation of the authors at the time of publication
Clinical Unit for Herd Health Management in Ruminants, University Clinic for Ruminants, Department for Farm Animals and Veterinary Public Health, University of Veterinary Medicine Vienna, Vienna, Austria;
Smartbow/Zoetis LLC, Weibern, Austria; 3Department of Clinical Sciences, Faculty of Veterinary Medicine, Ferdowsi University of Mashhad, Mashhad, Iran
Comments (0)
Ask the author
0 comments