Deep Learning for Laying Hen Activity Recognition Using Wearable Sensors
Laying hen activities in modern intensive housing systems can dramatically influence the policies needed for the optimal management of such systems. Intermittent monitoring of different behaviors during daytime cannot provide a good overview, since daily behaviors are not equally distributed over th...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Agriculture |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-0472/13/3/738 |
| _version_ | 1797614084077649920 |
|---|---|
| author | Mohammad Shahbazi Kamyar Mohammadi Sayed M. Derakhshani Peter W. G. Groot Koerkamp |
| author_facet | Mohammad Shahbazi Kamyar Mohammadi Sayed M. Derakhshani Peter W. G. Groot Koerkamp |
| author_sort | Mohammad Shahbazi |
| collection | DOAJ |
| description | Laying hen activities in modern intensive housing systems can dramatically influence the policies needed for the optimal management of such systems. Intermittent monitoring of different behaviors during daytime cannot provide a good overview, since daily behaviors are not equally distributed over the day. This paper investigates the application of deep learning technology in the automatic recognition of laying hen behaviors equipped with body-worn inertial measurement unit (IMU) modules in poultry systems. Motivated by the human activity recognition literature, a sophisticated preprocessing method is tailored on the time-series data of IMU, transforming it into the form of so-called activity images to be recognized by the deep learning models. The diverse range of behaviors a laying hen can exhibit are categorized into three classes: low-, medium-, and high-intensity activities, and various recognition models are trained to recognize these behaviors in real-time. Several ablation studies are conducted to assess the efficacy and robustness of the developed models against variations and limitations common for an in situ practical implementation. Overall, the best trained model on the full-feature acquired data achieves a mean accuracy of almost 100%, where the whole process of inference by the model takes less than 30 milliseconds. The results suggest that the application of deep learning technology for activity recognition of individual hens has the potential to accurately aid successful management of modern poultry systems. |
| first_indexed | 2024-03-11T07:04:23Z |
| format | Article |
| id | doaj.art-985c47892d7b4d2a9386209c623af1e3 |
| institution | Directory Open Access Journal |
| issn | 2077-0472 |
| language | English |
| last_indexed | 2024-03-11T07:04:23Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agriculture |
| spelling | doaj.art-985c47892d7b4d2a9386209c623af1e32023-11-17T09:02:48ZengMDPI AGAgriculture2077-04722023-03-0113373810.3390/agriculture13030738Deep Learning for Laying Hen Activity Recognition Using Wearable SensorsMohammad Shahbazi0Kamyar Mohammadi1Sayed M. Derakhshani2Peter W. G. Groot Koerkamp3School of Mechanical Engineering, Iran University of Science and Technology, Tehran 1684613114, IranSchool of Mechanical Engineering, Iran University of Science and Technology, Tehran 1684613114, IranWageningen Food and Biobased Research, 6700 AA Wageningen, The NetherlandsFarm Technology Group, Wageningen University, 6700 AA Wageningen, The NetherlandsLaying hen activities in modern intensive housing systems can dramatically influence the policies needed for the optimal management of such systems. Intermittent monitoring of different behaviors during daytime cannot provide a good overview, since daily behaviors are not equally distributed over the day. This paper investigates the application of deep learning technology in the automatic recognition of laying hen behaviors equipped with body-worn inertial measurement unit (IMU) modules in poultry systems. Motivated by the human activity recognition literature, a sophisticated preprocessing method is tailored on the time-series data of IMU, transforming it into the form of so-called activity images to be recognized by the deep learning models. The diverse range of behaviors a laying hen can exhibit are categorized into three classes: low-, medium-, and high-intensity activities, and various recognition models are trained to recognize these behaviors in real-time. Several ablation studies are conducted to assess the efficacy and robustness of the developed models against variations and limitations common for an in situ practical implementation. Overall, the best trained model on the full-feature acquired data achieves a mean accuracy of almost 100%, where the whole process of inference by the model takes less than 30 milliseconds. The results suggest that the application of deep learning technology for activity recognition of individual hens has the potential to accurately aid successful management of modern poultry systems.https://www.mdpi.com/2077-0472/13/3/738laying henactivity recognitionwearable sensordeep learningtime-series datasignal imaging |
| spellingShingle | Mohammad Shahbazi Kamyar Mohammadi Sayed M. Derakhshani Peter W. G. Groot Koerkamp Deep Learning for Laying Hen Activity Recognition Using Wearable Sensors Agriculture laying hen activity recognition wearable sensor deep learning time-series data signal imaging |
| title | Deep Learning for Laying Hen Activity Recognition Using Wearable Sensors |
| title_full | Deep Learning for Laying Hen Activity Recognition Using Wearable Sensors |
| title_fullStr | Deep Learning for Laying Hen Activity Recognition Using Wearable Sensors |
| title_full_unstemmed | Deep Learning for Laying Hen Activity Recognition Using Wearable Sensors |
| title_short | Deep Learning for Laying Hen Activity Recognition Using Wearable Sensors |
| title_sort | deep learning for laying hen activity recognition using wearable sensors |
| topic | laying hen activity recognition wearable sensor deep learning time-series data signal imaging |
| url | https://www.mdpi.com/2077-0472/13/3/738 |
| work_keys_str_mv | AT mohammadshahbazi deeplearningforlayinghenactivityrecognitionusingwearablesensors AT kamyarmohammadi deeplearningforlayinghenactivityrecognitionusingwearablesensors AT sayedmderakhshani deeplearningforlayinghenactivityrecognitionusingwearablesensors AT peterwggrootkoerkamp deeplearningforlayinghenactivityrecognitionusingwearablesensors |