Integrating UAV-based LiDAR and 3D computer vision to enhance cattle growth monitoring in precision livestock farming

Samenvatting

This thesis explores the use of Unmanned Aerial Vehicles (UAVs) equipped with a LiDAR device to monitor cattle growth in Precision Livestock Farming (PLF). The research specifically targets semi-free-range cattle rearing environments, where beef cattle are provided with controlled feeding and management while retaining some mobility within confined areas. The primary objective is to develop accurate, non-intrusive methodologies for cattle body measurement retrieval, individual identification, and body weight estimation. Unlike traditional methods such as manual operations and transponders, which pose risks to both animals and humans, or two-dimensional imaging techniques, which are often affected by environmental factors (such as illumination) and occlusions, the three-dimensional (3D) approach offers significant advantages. It provides depth data, overcoming the limitations of 2D methods. Using the high-precision remote sensing capabilities of UAV-based LiDAR systems, this thesis has demonstrated the significant potential of UAV-based LiDAR systems and advanced 3D CV techniques in revolutionising PLF. By addressing key challenges in individual cattle identification and body weight estimation for cattle with different postures, the research provides a robust framework for enhancing cattle management practices in semi-free-range environments. The methodologies developed here not only improve accuracy and efficiency in monitoring cattle growth but also offer scalable solutions adaptable for other species and broader agricultural contexts. As the field of PLF continues to evolve, the insights and innovations presented in this thesis serve as a foundation for future research, paving the way for more sophisticated, non-intrusive livestock management practices that prioritise animal welfare and operational efficiency.