Related Experiment Video
Updated: Dec 19, 2025
06:30
Noninvasive, In-pen Approach Test for Laboratory-housed Pigs
Published on: June 5, 2019
8.8K
Quantifying defence cascade responses as indicators of pig affect and welfare using computer vision methods.
Poppy Statham1, Sion Hannuna2, Samantha Jones1
1Animal Welfare and Behaviour Group, Bristol Veterinary School, University of Bristol, Langford House, Langford, BS40 5DU, UK.
Scientific Reports
|June 4, 2020
Summary
Computer vision accurately measures pig startle and freeze responses, offering a practical method for assessing animal welfare in field conditions. This technology aids in understanding affective states and improving animal well-being.
Area of Science:
- Animal Welfare Science
- Ethology
- Computer Vision Applications
Background:
- Affective states significantly influence animal welfare, making their assessment crucial.
- The rapid Defence Cascade (DC) response, including startle and freeze behaviors, is a potential indicator of negative affective states in animals.
- Field-based assessment of animal affect requires practical and easily measurable indicators.
Purpose of the Study:
- To evaluate the efficacy of computer vision in quantifying the Defence Cascade (DC) responses in pigs.
- To determine if computer vision can provide reliable measures of startle magnitude and freeze duration under field conditions.
- To explore the potential of computer vision as a tool for assessing pig affect and welfare.
Main Methods:
- Induced Defence Cascade (DC) responses were recorded in 12 pigs, generating 280 video clips.
- Ground truth measures of startle and freeze duration were established through manual video analysis.
- Computer vision image analysis (sparse feature tracking) was employed to estimate DC responses, alongside load platforms, Kinect depth cameras, and kinematic data.
Main Results:
- Computer vision image analysis data strongly predicted ground truth measures of DC responses.
- Image analysis data showed strong positive correlations with ground truth and all other measurement methods.
- Factors such as stimulus characteristics, pig orientation, and pre-response behavior influenced DC responses.
Conclusions:
- Computer vision offers a practical and reliable method for quantifying pig Defence Cascade (DC) responses.
- This approach has significant potential for assessing pig affect and welfare under field conditions.
- Computer vision technology can advance the field assessment of animal welfare.
