Uncovering Patterns in Dairy Cow Behaviour: A Deep Learning Approach with Tri-Axial Accelerometer Data

Abstract

The accurate detection of behavioural changes represents a promising method of detecting the early onset of disease in dairy cows. This study assessed the performance of deep learning (DL) in classifying dairy cows’ behaviour from accelerometry data acquired by single sensors on the cows’ left flanks and compared the results with those obtained through classical machine learning (ML) from the same raw data. Twelve cows with a tri-axial accelerometer were observed for 136 ± 29 min each to detect five main behaviours: standing still, moving, feeding, ruminating and resting. For each 8 s time interval, 15 metrics were calculated, obtaining a dataset of 211,720 observation units and 15 columns. The entire dataset was randomly split into training (80%) and testing (20%) datasets. The DL accuracy, precision and sensitivity/recall were calculated and compared with the performance of classical ML models. The best predictive model was an 8-layer convolutional neural network (CNN) with an overall accuracy and F1 score equal to 0.96. The precision, sensitivity/recall and F1 score of single behaviours had the following ranges: 0.93–0.99. The CNN outperformed all the classical ML algorithms. The CNN used to monitor the cows’ conditions showed an overall high performance in successfully predicting multiple behaviours using a single accelerometer.

Application of Precision Technologies to Characterize Animal Behavior: A Review

“…The combination of accelerometers and GPS results in a synergistic relationship that exploits the strengths of both sensors to provide a good understanding of ruminants. Australia Accuracy of 88% to 98% in monitoring licking behavior [42] Australia 4-month-old calves suckled fewer times, but for longer [73] United Kingdom Classification of rumination, eating, and other behaviors with precision of 0.83 [74] Pasture-based France The accuracy of prediction of the main behaviors was 98% [40] Semi-enclosed barn United States Accuracy of rumination detection was 86.2% [41] Three dairy farms Italy Accuracy of behavior detection was 85.12% [75] Dairy farm Italy Accuracy of classifying behavior was 96% [76] GPS Extensive United States Cattle followed water more than salt [3] Hungary Weather fronts affected the herd's route [64] Pasture-based Malaysia Observation of the grazing patterns was accurate [63] England Cattle tended to favor shorter material during the day and material of higher crude fiber in the evening [66] Commercial farm Spain Sensor was able to detect hotspots of dung deposition [77] GPS-GPRS Extensive Spain Distance traveled daily was 3147 m [65] Accelerometer, GPS Pasture-based Australia Description of the animals' movement and some behaviors was successful [78] Spain Accuracy of classification of behavior was 93% [70] Accelerometer, RFID Pasture-based Australia Accelerometer correlated highly with the observed duration of drinking events [79] Accelerometer, magnetometer Intensive Tasmania Grazing, ruminating, and resting were identified accurately [80] Accelerometer, cameras Intensive China Accuracy of 94.9% in recognizing behavior [81] Table 1. Cont.…”