Lambs' live weight estimation using 3D images.

Area of Science:

• Agricultural Science
• Animal Science
• Computer Vision

Background:

• The sheep sector lags in technological adoption compared to other livestock industries due to knowledge gaps and economic constraints.
• Precision Livestock Farming (PLF) offers potential efficiency improvements but requires accessible tools for sheep farmers.
• Accurate lamb weight estimation is crucial for farm management and profitability, yet current methods are imprecise and labor-intensive.

Purpose of the Study:

• To develop and evaluate a novel system for estimating lamb weight using 3D image capture.
• To address the challenges of inaccurate and time-consuming lamb weighing in sheep farming.
• To introduce an affordable and user-friendly technology to enhance Precision Livestock Farming practices in sheep production.

Main Methods:

• Recorded zenithal (top-down) 3D images of 272 lambs.
• Processed images by capturing the upper area and calculating the average pixel depth of each lamb.
• Developed a weight estimation algorithm based on these 3D image parameters.

Main Results:

• The proposed 3D imaging system estimated lamb weight with a mean error of less than 6%.
• The technology demonstrated low economic cost and ease of operation, encouraging farmer adoption.
• The method significantly reduced weighing process duration and animal stress while improving accuracy.

Conclusions:

• 3D image analysis provides a viable, accurate, and cost-effective solution for lamb weight estimation in sheep farming.
• This technology enhances Precision Livestock Farming by improving animal weight monitoring and management.
• Implementing this system can lead to better economic profitability for sheep farmers through improved weight control.