Integrated unmanned aerial vehicle-based LiDAR and RGB data for individual cattle growth monitoring in precision livestock farming
Abstract Monitoring cattle growth on an individual level is essential in precision livestock farming. Our study utilises an Unmanned Aerial Vehicle (UAV)-based Light Detection and Ranging (LiDAR) system to monitor the growth of 96 semi-free-range beef cattle. We conducted drone campaigns with varyin...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-07-01
|
| Series: | Scientific Data |
| Online Access: | https://doi.org/10.1038/s41597-025-04783-6 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Monitoring cattle growth on an individual level is essential in precision livestock farming. Our study utilises an Unmanned Aerial Vehicle (UAV)-based Light Detection and Ranging (LiDAR) system to monitor the growth of 96 semi-free-range beef cattle. We conducted drone campaigns with varying flight speeds and heights under different environmental conditions. The resulting point clouds were processed to segment individual cattle with both standing and lying postures. Concurrently, UAV-based RGB data underwent instance segmentation, categorising cattle images at an individual level. In addition, ground-based cameras recorded the cattle weighing process. Video frames were extracted, selected, segmented, and stored, providing diverse views of each animal. These individual point clouds were processed, resulting in the extracted body measurements (such as height, width, and length), and volumes of point clouds. These results can facilitate the evaluation of the live body weight of cattle, corroborated with the ground truth data provided. Segmented RGB data can support the identification of cattle based on biometric characteristics, such as coat patterns. This comprehensive dataset holds significant potential for research in cattle monitoring using three-dimensional methods. |
|---|---|
| ISSN: | 2052-4463 |