The effect of different time epoch settings on the classification of sheep behaviour using tri-axial accelerometry

Decandia, M.; Giovanetti, V.; Molle, G.; Acciaro, M.; Mameli, M.; Cabiddu, A.; Cossu, R.; Serra, Moisès Serra i; Manca, C.; Rassu, S.P.G.; Dimauro, Corrado

doi:10.1016/j.compag.2018.09.002

Cited by 46 publications

(33 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the results, we noted that 5-second windows can provide a very good activity pattern representation and therefore could be suggested that this size is adequate. However, Decandia et al [32], conducted experiments with various window sizes, such as 5, 10, 30, 60, 120, 180 and 300 seconds, and they identified that the best performance was obtained from a 30 second window having sensitivity 94.8% for grazing, 80.4% for ruminating, and 92.3% for other behaviours. Though, the two studies cannot be compared because the ML model applied, the selection of features, and the position of the sensor is different.…”

Section: Resultsmentioning

confidence: 99%

“…Decandia et al [32] evaluated the performance of canonical discriminant analysis (CDA), and discriminant analysis (DA) to distinguish between three behaviours of sheep; grazing, ruminating, and others. The authors aimed to identify the window which provides the best algorithm performance and they evaluated windows of 5, 10, 30, 60, 120, 180 and 300 s from accelerometer signals sampled at 62.5Hz.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Feature Extraction and Random Forest to Identify Sheep Behavior from Accelerometer Data

Kleanthous

Hussain

Khan

et al. 2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Sensor technologies play an essential part in the agricultural community and many other scientific and commercial communities. Accelerometer signals and Machine Learning techniques can be used to identify and observe behaviours of animals without the need for an exhaustive human observation which is labour intensive and time consuming. This study employed random forest algorithm to identify grazing, walking, scratching, and inactivity (standing, resting) of 8 Hebridean ewes located in Cheshire, Shotwick in the UK. We gathered accelerometer data from a sensor device which was fitted on the collar of the animals. The selection of the algorithm was based on previous research by which random forest achieved the best results among other benchmark techniques. Therefore, in this study, more focus was given to feature engineering to improve prediction performance. Seventeen features from time and frequency domain were calculated from the accelerometer measurements and the magnitude of the acceleration. Feature elimination was utilised in which highly correlated ones were removed, and only nine out of seventeen features were selected. The algorithm achieved an overall accuracy of 99.43% and a kappa value of 98.66%. The accuracy for grazing, walking, scratching, and inactive was 99.08%, 99.13%, 99.90%, and 99.85%, respectively. The overall results showed that there is a significant improvement over previous methods and studies for all mutually exclusive behaviours. Those results are promising, and the technique could be further tested for future real-time activity recognition.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Feature Extraction and Random Forest to Identify Sheep Behavior from Accelerometer Data

Kleanthous

Hussain

Khan

et al. 2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Limiting the need to process, store, and transmit data, will extend battery life and ultimately reduce costs. Decandia et al (2018) reported a similar study assessing sheep behavior classification. They found that aggregating to 30 s was most accurate having tested aggregations from 5 to 300 s. The collected data and the reference or gold standard also need to be aligned.…”

Section: System Designmentioning

confidence: 96%

Invited review: Cattle lameness detection with accelerometers

O’Leary

Byrne

O’Connor

et al. 2020

Journal of Dairy Science

View full text Add to dashboard Cite

“…These tags measure linear acceleration along one or more axes and can accurately depict animal body movement [102]. Additionally, the sampling frequency can influence the accuracy of identifying a certain behaviour, for instance grazing, and should be carefully selected [103]. Accelerometers consist of integrated data loggers and can be attached to various parts of the body, depending on the behaviours which are being observed [104,112] (Fig.…”

Section: Accelerometersmentioning

confidence: 99%

Technologies for the automated collection of heat stress data in sheep

et al. 2021

View full text Add to dashboard Cite

The automated collection of phenotypic measurements in livestock is becoming increasingly important to both researchers and farmers. The capacity to non-invasively collect real-time data, provides the opportunity to better understand livestock behaviour and physiology and improve animal management decisions. Current climate models project that temperatures will increase across the world, influencing both local and global agriculture. Sheep that are exposed to high ambient temperatures experience heat stress and their physiology, reproductive function and performance are compromised. Body temperature is a reliable measure of heat stress and hence a good indicator of an animals’ health and well-being. Non-invasive temperature-sensing technologies have made substantial progress over the past decade. Here, we review the different technologies available and assess their suitability for inferring ovine heat stress. Specifically, the use of indwelling probes, intra-ruminal bolus insertion, thermal imaging and implantable devices are investigated. We further evaluate the capacity of behavioural tracking technology, such as global positioning systems, to identify heat stressed individuals based on the exhibition of specific behaviours. Although there are challenges associated with using real-time thermosensing data to make informed management decisions, these technologies provide new opportunities to manage heat stress in sheep. In order to obtain accurate real-time information of individual animals and facilitate prompt intervention, data collection should be entirely automated. Additionally, for accurate interpretation on-farm, the development of software which can effectively collect, manage and integrate data for sheep producer’s needs to be prioritised. Lastly, understanding known physiological thresholds will allow farmers to determine individual heat stress risk and facilitate early intervention to reduce the effects in both current and subsequent generations.

show abstract

The effect of different time epoch settings on the classification of sheep behaviour using tri-axial accelerometry

Cited by 46 publications

References 24 publications

Feature Extraction and Random Forest to Identify Sheep Behavior from Accelerometer Data

Feature Extraction and Random Forest to Identify Sheep Behavior from Accelerometer Data

Invited review: Cattle lameness detection with accelerometers

Technologies for the automated collection of heat stress data in sheep

Contact Info

Product

Resources

About