Scholarlink

<p>Manually auditing gait scores of broiler chickens is labor-intensive and subjective, necessitating the development of an automated and objective alternative. This study presents a novel three-dimensional (3D) deep learning pipeline to assess the walking ability of broilers by predicting gait scores ranging from 0 (optimal mobility) to 2 (severely impaired mobility). A total of 540 broiler chicken videos, sampled from 6 to 7 weeks of age, were recorded as the chickens traversed a 1.75-meter wooden platform. An Intel RealSense L515 LiDAR camera, mounted at a 2.5-meter height, was used to capture synchronized RGB-Depth data. The data were recorded using the Robot Operating System (ROS) Noetic, ensuring efficient and structured data acquisition. The proposed pipeline consisted of multiple sequential steps: (1) RGB-Depth frame extraction and synchronization, (2) pose estimation using a custom-trained YOLOv11-based chicken pose detection model, (3) back-projection of 2D keypoints into 3D space using camera intrinsics, (4) frame validation using a convolutional neural network to filter out occlusions and artifacts, (5) platform orientation detection via Hough line transformation, (6) segmentation of the chicken’s body using the Segment Anything Model (SAM) to extract 3D point clouds, and (7) kinematic feature extraction for gait analysis. Key 3D features, including velocity, acceleration, and head turn frequency, were fed into a multi-layer perceptron classifier to predict gait scores. The classifier predicted broiler gait scores with 93.34% accuracy, 95.56% precision, 91.16% recall, and 93.31% F1-score. The system demonstrated robustness in various environmental conditions, offering a scalable and cost-effective solution for poultry welfare monitoring. The entire system was developed at an approximate cost of $1483, making it an affordable alternative for large-scale poultry research and field sampling operations. This research highlights the potential of integrating 3D vision with deep learning to enhance automated gait scoring, ultimately improving assessment reliability, reducing labor costs, and contributing to the advancement of precision poultry farming.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A novel three-dimensional deep learning approach for auditing gait scores of individual broiler chickens

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月12日

查看全文
  • Ehsan Asali,
  • Guoming Li,
  • Tongshuai Liu,
  • Chongxiao Chen,
  • Mahtab Saeidifar,
  • Venkat Umesh Chandra Bodempudi,
  • Oluwadamilola Moyin Oso,
  • Aravind Mandiga,
  • Sai Akshitha Reddy Kota,
  • Tianming Liu

摘要

Manually auditing gait scores of broiler chickens is labor-intensive and subjective, necessitating the development of an automated and objective alternative. This study presents a novel three-dimensional (3D) deep learning pipeline to assess the walking ability of broilers by predicting gait scores ranging from 0 (optimal mobility) to 2 (severely impaired mobility). A total of 540 broiler chicken videos, sampled from 6 to 7 weeks of age, were recorded as the chickens traversed a 1.75-meter wooden platform. An Intel RealSense L515 LiDAR camera, mounted at a 2.5-meter height, was used to capture synchronized RGB-Depth data. The data were recorded using the Robot Operating System (ROS) Noetic, ensuring efficient and structured data acquisition. The proposed pipeline consisted of multiple sequential steps: (1) RGB-Depth frame extraction and synchronization, (2) pose estimation using a custom-trained YOLOv11-based chicken pose detection model, (3) back-projection of 2D keypoints into 3D space using camera intrinsics, (4) frame validation using a convolutional neural network to filter out occlusions and artifacts, (5) platform orientation detection via Hough line transformation, (6) segmentation of the chicken’s body using the Segment Anything Model (SAM) to extract 3D point clouds, and (7) kinematic feature extraction for gait analysis. Key 3D features, including velocity, acceleration, and head turn frequency, were fed into a multi-layer perceptron classifier to predict gait scores. The classifier predicted broiler gait scores with 93.34% accuracy, 95.56% precision, 91.16% recall, and 93.31% F1-score. The system demonstrated robustness in various environmental conditions, offering a scalable and cost-effective solution for poultry welfare monitoring. The entire system was developed at an approximate cost of $1483, making it an affordable alternative for large-scale poultry research and field sampling operations. This research highlights the potential of integrating 3D vision with deep learning to enhance automated gait scoring, ultimately improving assessment reliability, reducing labor costs, and contributing to the advancement of precision poultry farming.