Related Experiment Video
Updated: Nov 15, 2025
06:19
Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction
Published on: August 16, 2024
669
Practices and Applications of Convolutional Neural Network-Based Computer Vision Systems in Animal Farming: A Review
Guoming Li1, Yanbo Huang2, Zhiqian Chen3
1Department of Agricultural and Biological Engineering, Mississippi State University, Starkville, MS 39762, USA.
Sensors (Basel, Switzerland)
|March 6, 2021
Summary
Convolutional neural networks (CNNs) enhance farm animal management through computer vision. This review details CNN applications, challenges, and future directions for improved animal welfare and productivity.
Area of Science:
- Computer Vision in Agriculture
- Deep Learning for Animal Science
Background:
- Convolutional neural networks (CNNs) are increasingly used in animal farming for management.
- Existing knowledge on CNN applications, practices, limitations, and solutions requires expansion.
Purpose of the Study:
- To systematically review CNN-based computer vision applications in animal farming.
- To cover five key deep learning tasks: image classification, object detection, segmentation, pose estimation, and tracking.
Main Methods:
- Reviewed CNN architectures and algorithm development strategies for animal farming.
- Summarized system development preparations, including data recording and preprocessing.
- Organized system applications by year, country, animal species, and purpose.
Main Results:
- Identified cattle, sheep/goats, pigs, and poultry as major farm animal species of concern.
- Discussed model performance, optimization practices, and CNN architectures.
- Highlighted system applications and provided recommendations for future research.
Conclusions:
- CNN-based computer vision offers significant potential for advancing animal farming.
- Future research should focus on improving CNN systems for enhanced animal welfare, environment, engineering, genetics, and management.
