Chinese Academy of Sciences Researchers Develop Efficient Algorithm for Small Livestock Object Detection in Unmanned Aerial Vehicle Imagery

Strategic Insights -

Researchers from the Chinese Academy of Sciences have made a significant breakthrough in developing an efficient algorithm for small livestock object detection in unmanned aerial vehicle (UAV) imagery. Their novel approach, known as Livestock Network (LSNET), incorporates a low-level prediction head and a Large Kernel Attentions Spatial Pyramid Pooling (LKASPP) module to detect small livestock objects in expansive farms. The researchers claim that their approach overcomes the limitations of existing methods and contributes to more effective livestock management and advancements in agricultural technology.

Key Takeaways:

  • The researchers developed a novel algorithm, Livestock Network (LSNET), for small livestock object detection in UAV imagery, improving mean Average Precision (mAP) by 1.47% compared to YOLOv7.
  • The LSNET algorithm incorporates a low-level prediction head (P2) and a Large Kernel Attentions Spatial Pyramid Pooling (LKASPP) module to detect small livestock objects.
  • The research addresses the limitations of existing methods, including small and densely packed livestock objects in UAV images, by introducing a new YOLOv7-based network.
  • The study used a dataset of grazing livestock from the Prairie Chenbarhu Banner in Hulunbuir, Inner Mongolia, collected using UAV images.
  • The researchers claim that their approach offers a practical solution for livestock detection in expansive farms.
  • The research was funded by The National Key R&D Program of China, Strategic Priority Research Program of The Chinese Academy of Sciences, Science And Technology Program of Tianjin, and The Key Laboratory For Digital Land And Resources of Jiangxi Province.

Statistics:

  • 1.47% improvement in mean Average Precision (mAP) achieved by the proposed LSNET algorithm compared to YOLOv7.
  • The dataset used in the study covered grazing livestock from the Prairie Chenbarhu Banner in Hulunbuir, Inner Mongolia.
  • The research was conducted by Wenbo Chen and Dongliang Wang, in collaboration with Xiaowei Xie.

Sources:

  • Wenbo Chen, et al. An Efficient Algorithm for Small Livestock Object Detection in Unmanned Aerial Vehicle Imagery. Animals, 2025,15(12):1794.
  • NewsRx. Chinese Academy of Sciences Researchers Further Understanding of Livestock (An Efficient Algorithm for Small Livestock Object Detection in Unmanned Aerial Vehicle Imagery). Agriculture Week. July 10, 2025; p 41.