Counting Honey Bees Using Video Tracking
Author Information
Author(s): Lei Chaokai, Lu Yuntao, Xing Zhiyuan, Zhang Jie, Li Shijuan, Wu Wei, Liu Shengping, Gray Alison
Primary Institution: Key Laboratory of Agricultural Blockchain Application, Agricultural Information Institute, Chinese Academy of Agricultural Sciences, Ministry of Agriculture and Rural Affairs, Beijing, China
Hypothesis
Can a video-based counting method improve the efficiency of monitoring honey bee in-and-out activity?
Conclusion
The study developed a video-based counting model that effectively tracks and counts honey bees, achieving high accuracy.
Supporting Evidence
- The model achieved F1in of 91.49% and F1out of 89.08%.
- YOLOv8m was identified as the best object detection model.
- OC-SORT was the most effective tracking algorithm for this study.
- The box method improved counting accuracy compared to the single-line method.
Takeaway
This study created a way to count how many bees go in and out of a hive using cameras and smart technology, making it easier for beekeepers.
Methodology
The study used video data from smart beehives and applied object detection and multiple object tracking algorithms to count bee movements.
Limitations
The counting method may still be affected by identity switches among bees and environmental factors.
Participant Demographics
The study focused on Italian honey bees (Apis mellifera) in Beijing, China.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website