Related Experiment Video
Updated: Sep 27, 2025
08:32
Tracking Rats in Operant Conditioning Chambers Using a Versatile Homemade Video Camera and DeepLabCut
Published on: June 15, 2020
12.7K
EmbeddedPigCount: Pig Counting with Video Object Detection and Tracking on an Embedded Board.
Jonggwan Kim1, Yooil Suh1, Junhee Lee1
1Info Valley Korea Co., Ltd., Anyang-si 14067, Korea.
Sensors (Basel, Switzerland)
|April 12, 2022
Summary
Accurately counting pigs on farms is challenging. This study introduces an automated, camera-based deep learning method for real-time pig counting, achieving 99.44% accuracy on embedded systems.
Area of Science:
- Agricultural Technology
- Computer Vision
- Animal Science
Background:
- Accurate livestock counting is crucial for efficient farm management.
- Manual pig counting is labor-intensive and prone to errors due to animal movement.
Purpose of the Study:
- To develop an automated, real-time pig counting system using camera-based deep learning.
- To optimize the method for low-cost embedded systems, balancing accuracy and speed.
Main Methods:
- Implemented a deep learning model for video object detection and tracking.
- Utilized a camera in a hallway to analyze pig traffic.
- Optimized the algorithm for real-time performance on an NVIDIA Jetson Nano board.
Main Results:
- Achieved a high counting accuracy of 99.44%.
- Demonstrated effective real-time execution on an embedded system.
- The method successfully counted pigs even with bidirectional movement.
Conclusions:
- The proposed lightweight, deep learning-based method provides an accurate and efficient solution for automated pig counting.
- This technology can significantly improve farm management efficiency and reduce labor costs.
