Implementation of Computer-Vision-Based Farrowing Prediction in Pens with Temporary Sow Confinement

Simple Summary The automated detection of the beginning and ending of nest-building behavior in sows might support the implementation of new management strategies in farrowing pens, i.e., temporary sow confinement in crates. This might improve sows’ welfare and reduce piglet mortality by limiting sow confinement in crates to only a critical period of piglets’ lives. The objective of this study was to predict farrowing with computer vision techniques to optimize the timing of sow confinement. In this study, we developed a computer-vision-based method for the automated detection of the beginning and ending of nest-building behavior in sows. The study included 71 Austrian Large White and Landrace × Large White crossbred sows and four types of farrowing pens. The beginning of nest-building behavior was detected with a median of 12 h 51 min and ending with a median of 2 h 38 min before the beginning of farrowing. It was possible to predict farrowing for 29 out of 44 animals. The developed method could be applied to warn the farmer when nest-building behavior starts and then to confine the sow in a crate when the end of nest-building behavior is detected. This could reduce labor costs otherwise required for the regular control of sows in farrowing compartments. Abstract The adoption of temporary sow confinement could improve animal welfare during farrowing for both the sow and the piglets. An important challenge related to the implementation of temporary sow confinement is the optimal timing of confinement in crates, considering sow welfare and piglet survival. The objective of this study was to predict farrowing with computer vision techniques to optimize the timing of sow confinement. In total, 71 Austrian Large White and Landrace × Large White crossbred sows and four types of farrowing pens were included in the observational study. We applied computer vision model You Only Look Once X to detect sow locations, the calculated activity level of sows based on detected locations and detected changes in sow activity trends with Kalman filtering and the fixed interval smoothing algorithm. The results indicated the beginning of nest-building behavior with a median of 12 h 51 min and ending with a median of 2 h 38 min before the beginning of farrowing with the YOLOX-large object detection model. It was possible to predict farrowing for 29 out of 44 sows. The developed method could reduce labor costs otherwise required for the regular control of sows in farrowing compartments.

[1]  M. Oczak,et al.  Where is the sow’s nose: RetinaNet object detector as a basis for monitoring the use of rack with nest-building material , 2022, Frontiers in Animal Science.

[2]  Shiliang Zhang,et al.  Contextual Instance Decoupling for Robust Multi-Person Pose Estimation , 2022, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  S. Edwards,et al.  Review of Temporary Crating of Farrowing and Lactating Sows , 2022, Frontiers in Veterinary Science.

[4]  M. Oczak,et al.  Comparison of the automated monitoring of the sow activity in farrowing pens using video and accelerometer data , 2022, Comput. Electron. Agric..

[5]  Bernd Stock,et al.  Automatic Behavior and Posture Detection of Sows in Loose Farrowing Pens Based on 2D-Video Images , 2021, Frontiers in Animal Science.

[6]  Zeming Li,et al.  YOLOX: Exceeding YOLO Series in 2021 , 2021, ArXiv.

[7]  Sujoy Ganguly,et al.  Unity Perception: Generate Synthetic Data for Computer Vision , 2021, ArXiv.

[8]  T. Rodenburg,et al.  Individual Detection and Tracking of Group Housed Pigs in Their Home Pen Using Computer Vision , 2021, Frontiers in Animal Science.

[9]  European Citizens ' Initiative " End the cage age " , 2021 .

[10]  Boris Sekachev,et al.  opencv/cvat: v1.1.0 , 2020 .

[11]  Xiangyu Zhang,et al.  Learning Delicate Local Representations for Multi-Person Pose Estimation , 2020, ECCV.

[12]  Ross B. Girshick,et al.  Focal Loss for Dense Object Detection , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  M. Oczak,et al.  Dynamics of Sows’ Activity Housed in Farrowing Pens with Possibility of Temporary Crating might Indicate the Time When Sows Should be Confined in a Crate before the Onset of Farrowing , 2019, Animals : an open access journal from MDPI.

[14]  Talmo D. Pereira,et al.  Fast animal pose estimation using deep neural networks , 2018, bioRxiv.

[15]  Ali Farhadi,et al.  YOLOv3: An Incremental Improvement , 2018, ArXiv.

[16]  Chen Chen,et al.  Image motion feature extraction for recognition of aggressive behaviors among group-housed pigs , 2017, Comput. Electron. Agric..

[17]  Zhanxing Zhu,et al.  Towards Understanding Generalization of Deep Learning: Perspective of Loss Landscapes , 2017, ArXiv.

[18]  Dries Berckmans,et al.  General introduction to precision livestock farming , 2017 .

[19]  S. Jarvis,et al.  The identification of potential behavioural indicators of pain in periparturient sows. , 2016, Research in veterinary science.

[20]  Marco Gamba,et al.  BORIS: a free, versatile open‐source event‐logging software for video/audio coding and live observations , 2016 .

[21]  Matti Pastell,et al.  Predicting farrowing of sows housed in crates and pens using accelerometers and CUSUM charts , 2016, Comput. Electron. Agric..

[22]  Yongwha Chung,et al.  Automatic Recognition of Aggressive Behavior in Pigs Using a Kinect Depth Sensor , 2016, Sensors.

[23]  Maciej Oczak,et al.  Classification of nest-building behaviour in non-crated farrowing sows on the basis of accelerometer data , 2015 .

[24]  C. Lindegaard,et al.  An equine pain face , 2014, Veterinary anaesthesia and analgesia.

[25]  Pietro Perona,et al.  Microsoft COCO: Common Objects in Context , 2014, ECCV.

[26]  Jukka Heikkonen,et al.  Using movement sensors to detect the onset of farrowing , 2008 .

[27]  Y Koketsu,et al.  Variability and repeatability in gestation length related to litter performance in female pigs on commercial farms. , 2007, Theriogenology.

[28]  Peter C. Young,et al.  THE CAPTAIN TOOLBOX FOR MATLAB , 2006 .

[29]  B Erez,et al.  A microcomputer-photocell system to monitor periparturient activity of sows and transfer data to remote location. , 1990, Journal of animal science.