Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped Collars
Monitoring and classification of dairy cattle behaviours is essential for optimising milk yields. Early detection of illness, days before the critical conditions occur, together with automatic detection of the onset of oestrus cycles is crucial for obviating prolonged cattle treatments and improving...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/6/2323 |
| _version_ | 1797442319317729280 |
|---|---|
| author | Dejan Pavlovic Mikolaj Czerkawski Christopher Davison Oskar Marko Craig Michie Robert Atkinson Vladimir Crnojevic Ivan Andonovic Vladimir Rajovic Goran Kvascev Christos Tachtatzis |
| author_facet | Dejan Pavlovic Mikolaj Czerkawski Christopher Davison Oskar Marko Craig Michie Robert Atkinson Vladimir Crnojevic Ivan Andonovic Vladimir Rajovic Goran Kvascev Christos Tachtatzis |
| author_sort | Dejan Pavlovic |
| collection | DOAJ |
| description | Monitoring and classification of dairy cattle behaviours is essential for optimising milk yields. Early detection of illness, days before the critical conditions occur, together with automatic detection of the onset of oestrus cycles is crucial for obviating prolonged cattle treatments and improving the pregnancy rates. Accelerometer-based sensor systems are becoming increasingly popular, as they are automatically providing information about key cattle behaviours such as the level of restlessness and the time spent ruminating and eating, proxy measurements that indicate the onset of heat events and overall welfare, at an individual animal level. This paper reports on an approach to the development of algorithms that classify key cattle states based on a systematic dimensionality reduction process through two feature selection techniques. These are based on Mutual Information and Backward Feature Elimination and applied on knowledge-specific and generic time-series extracted from raw accelerometer data. The extracted features are then used to train classification models based on a Hidden Markov Model, Linear Discriminant Analysis and Partial Least Squares Discriminant Analysis. The proposed feature engineering methodology permits model deployment within the computing and memory restrictions imposed by operational settings. The models were based on measurement data from 18 steers, each animal equipped with an accelerometer-based neck-mounted collar and muzzle-mounted halter, the latter providing the truthing data. A total of 42 time-series features were initially extracted and the trade-off between model performance, computational complexity and memory footprint was explored. Results show that the classification model that best balances performance and computation complexity is based on Linear Discriminant Analysis using features selected through Backward Feature Elimination. The final model requires 1.83 ± 1.00 ms to perform feature extraction with 0.05 ± 0.01 ms for inference with an overall balanced accuracy of 0.83. |
| first_indexed | 2024-03-09T12:40:07Z |
| format | Article |
| id | doaj.art-b4ff19074643401a987e19086c05c554 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T12:40:07Z |
| publishDate | 2022-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-b4ff19074643401a987e19086c05c5542023-11-30T22:19:23ZengMDPI AGSensors1424-82202022-03-01226232310.3390/s22062323Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped CollarsDejan Pavlovic0Mikolaj Czerkawski1Christopher Davison2Oskar Marko3Craig Michie4Robert Atkinson5Vladimir Crnojevic6Ivan Andonovic7Vladimir Rajovic8Goran Kvascev9Christos Tachtatzis10BioSense Institute, 21101 Novi Sad, SerbiaDepartment of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1RD, UKDepartment of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1RD, UKBioSense Institute, 21101 Novi Sad, SerbiaDepartment of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1RD, UKDepartment of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1RD, UKBioSense Institute, 21101 Novi Sad, SerbiaDepartment of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1RD, UKSchool of Electrical Engineering, University of Belgrade, 11000 Belgrade, SerbiaSchool of Electrical Engineering, University of Belgrade, 11000 Belgrade, SerbiaDepartment of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1RD, UKMonitoring and classification of dairy cattle behaviours is essential for optimising milk yields. Early detection of illness, days before the critical conditions occur, together with automatic detection of the onset of oestrus cycles is crucial for obviating prolonged cattle treatments and improving the pregnancy rates. Accelerometer-based sensor systems are becoming increasingly popular, as they are automatically providing information about key cattle behaviours such as the level of restlessness and the time spent ruminating and eating, proxy measurements that indicate the onset of heat events and overall welfare, at an individual animal level. This paper reports on an approach to the development of algorithms that classify key cattle states based on a systematic dimensionality reduction process through two feature selection techniques. These are based on Mutual Information and Backward Feature Elimination and applied on knowledge-specific and generic time-series extracted from raw accelerometer data. The extracted features are then used to train classification models based on a Hidden Markov Model, Linear Discriminant Analysis and Partial Least Squares Discriminant Analysis. The proposed feature engineering methodology permits model deployment within the computing and memory restrictions imposed by operational settings. The models were based on measurement data from 18 steers, each animal equipped with an accelerometer-based neck-mounted collar and muzzle-mounted halter, the latter providing the truthing data. A total of 42 time-series features were initially extracted and the trade-off between model performance, computational complexity and memory footprint was explored. Results show that the classification model that best balances performance and computation complexity is based on Linear Discriminant Analysis using features selected through Backward Feature Elimination. The final model requires 1.83 ± 1.00 ms to perform feature extraction with 0.05 ± 0.01 ms for inference with an overall balanced accuracy of 0.83.https://www.mdpi.com/1424-8220/22/6/2323precision agriculturecattle behaviour monitoringfeature selection |
| spellingShingle | Dejan Pavlovic Mikolaj Czerkawski Christopher Davison Oskar Marko Craig Michie Robert Atkinson Vladimir Crnojevic Ivan Andonovic Vladimir Rajovic Goran Kvascev Christos Tachtatzis Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped Collars Sensors precision agriculture cattle behaviour monitoring feature selection |
| title | Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped Collars |
| title_full | Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped Collars |
| title_fullStr | Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped Collars |
| title_full_unstemmed | Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped Collars |
| title_short | Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped Collars |
| title_sort | behavioural classification of cattle using neck mounted accelerometer equipped collars |
| topic | precision agriculture cattle behaviour monitoring feature selection |
| url | https://www.mdpi.com/1424-8220/22/6/2323 |
| work_keys_str_mv | AT dejanpavlovic behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars AT mikolajczerkawski behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars AT christopherdavison behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars AT oskarmarko behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars AT craigmichie behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars AT robertatkinson behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars AT vladimircrnojevic behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars AT ivanandonovic behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars AT vladimirrajovic behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars AT gorankvascev behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars AT christostachtatzis behaviouralclassificationofcattleusingneckmountedaccelerometerequippedcollars |