Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Regulation of Goat Milk Protein Synthesis: Genetic Architecture, Signalling Pathways, and Omics Insights.

Molecular biotechnology·2026
Same author

Assessment of deep morphology and cryodamage in spermatozoa of indigenous and exotic pig breeds from the Northeast Indian Himalayas using scanning electron microscopy (SEM).

Open veterinary journal·2026
Same author

Expression profile of MHC class-II trans activator gene in pig immunized with recombinant virus like particles encoding porcine circovirus 2d.

Microbial pathogenesis·2026
Same author

Metabolomics and Artificial Intelligence (AI) assisted metabolomics in the diagnosis of non-tuberculous mycobacterial infections: progress and challenges.

New microbes and new infections·2026
Same author

First Molecular Detection of Japanese Encephalitis Virus Genotype I in Culex Mosquitoes from Assam, Northeast India.

The American journal of tropical medicine and hygiene·2026
Same author

Recent Applications of Tissue-Engineered 3D Scaffolds in Oncology: Present and Future Perspectives.

Protein and peptide letters·2026

Related Experiment Video

Updated: Jun 21, 2025

Noninvasive, In-pen Approach Test for Laboratory-housed Pigs
06:30

Noninvasive, In-pen Approach Test for Laboratory-housed Pigs

Published on: June 5, 2019

8.4K

Analysis of pig posture detection in group-housed pigs using deep learning-based mask scoring instance segmentation.

Salam Jayachitra Devi1, Juwar Doley1, Jaya Bharati1

  • 1ICAR-National Research Centre on Pig, Rani, Guwahati, Assam, India.

Animal Science Journal = Nihon Chikusan Gakkaiho
|July 15, 2024
PubMed
Summary

This study introduces a deep learning algorithm for accurate pig posture detection in group settings, improving livestock welfare monitoring. The method achieves over 96% accuracy in identifying individual pig postures like lying and walking.

Keywords:
artificial intelligencegroup‐housed piginstance segmentationpig posture

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Tracking Rats in Operant Conditioning Chambers Using a Versatile Homemade Video Camera and DeepLabCut
08:32

Tracking Rats in Operant Conditioning Chambers Using a Versatile Homemade Video Camera and DeepLabCut

Published on: June 15, 2020

12.4K

Related Experiment Videos

Last Updated: Jun 21, 2025

Noninvasive, In-pen Approach Test for Laboratory-housed Pigs
06:30

Noninvasive, In-pen Approach Test for Laboratory-housed Pigs

Published on: June 5, 2019

8.4K
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Tracking Rats in Operant Conditioning Chambers Using a Versatile Homemade Video Camera and DeepLabCut
08:32

Tracking Rats in Operant Conditioning Chambers Using a Versatile Homemade Video Camera and DeepLabCut

Published on: June 15, 2020

12.4K

Area of Science:

  • Animal Science
  • Computer Vision
  • Machine Learning

Background:

  • Pig posture is a key indicator of livestock health and welfare.
  • Deep learning for pig posture detection faces challenges with image variations and multiple pigs.

Purpose of the Study:

  • To develop and evaluate an instance segmentation algorithm for detecting and segmenting individual pig postures in group images.
  • To enable precise identification of postures including sternal lying, lateral lying, walking, and sitting.

Main Methods:

  • Utilized a ResNet-50 and Feature Pyramid Network for feature extraction.
  • Employed a region candidate network for region of interest (RoI) generation.
  • Applied non-maximum suppression (NMS) with a 0.7 threshold to handle overlapping pigs and improve detection accuracy.

Main Results:

  • Achieved over 96% accuracy in pig posture detection.
  • Reached a mean average precision (mAP) exceeding 83% for object detection and instance segmentation.
  • Identified optimal hyperparameters including a learning rate of 0.01 and a batch size of 512.

Conclusions:

  • The proposed instance segmentation algorithm effectively detects and segments individual pig postures in group images.
  • The method demonstrates high accuracy and robustness, outperforming comparisons like faster R-CNN.
  • This technology offers significant potential for enhanced livestock welfare monitoring.