Short communication: Measuring feed volume and weight by machine vision.

Area of Science:

• Agricultural Engineering
• Animal Science
• Machine Vision

Background:

• Individual dairy cow feed intake is crucial for health and productivity.
• Current feed monitoring systems are labor-intensive and costly.
• Automated systems can enhance dairy farm management.

Purpose of the Study:

• To evaluate an inexpensive 3D video camera system for monitoring dairy cow feed intake.
• To determine the accuracy of estimating feed weight from measured feed volume.
• To assess the system's potential for non-disruptive, automated feed monitoring.

Main Methods:

• Utilized a 3D video camera to measure feed volume.
• Derived feed weight from volume measurements.
• Performed regression analysis (linear and quadratic least squares t-test) on volume and weight data.
• Examined effects of feed positioning and sensor limitations.

Main Results:

• Accurate estimation of feed weight from 3D volume scans was achieved, with an error of less than 0.5 kg compared to digital scale measurements.
• The system demonstrated effectiveness in proof-of-concept testing.
• The method proved capable of non-disruptive feed intake monitoring.

Conclusions:

• An inexpensive 3D machine vision system can reliably estimate dairy cow feed weight.
• This technology offers a cost-effective and efficient alternative to traditional monitoring methods.
• Future work will involve real-world application in active bunks and with varied feed compositions.