Estimation of pig weight using a Microsoft Kinect prototype imaging system

• A Kinect prototype for estimation the weight of pigs is presented.
• The depth map images from Kinect adds information about animal height in addition to area.
• Manual intervention with respect to image quality is needed in prototype method.
• The accuracy of the Kinect prototype is 4–5% of mean weight.

The weight or mass of a pig is of great importance for farmers and stockmen to monitor performance, health and market weight of animals. The paper presents a prototype for pig weighing based on the Microsoft Kinect camera technology, utilizing the infrared depth map images. The system successfully estimated the weight of two different purebred breeds, landrace and duroc with an error estimate of 4–5% of mean weight. The depth map images require less calibration, are less prone to background (i.e. floor) noise compared to visible light camera systems and seem to be more robust between breeds due to additional information from height (depth map) of animals.

Figure optionsDownload as PowerPoint slide