Abstract

The Lipizzan horse is one of the oldest European horse breeds, its documented origins date back to 1580 in the Imperial stud-farm at Lipizza and are currently used the high-riding school, dressage and others equestrian sports as well as recreational activities. They are also traditionally selected by their morphological characteristics. The aim of this study was to verify the accuracy between manual measurements and stereo-image ones, for the most important linear, angular and circumference measures used for breeding purposes. In this approach a dual web-camera system, in combination with an image analysis algorithm, is proposed to automatically extract the information needed. For this reason a model species such as Lipizzan horse has been chosen and the most important biometric variables (linear measurements and angles) have been selected and taken into consideration. Ten horses were analyzed manually and using a stereovision system taking into account six linear measurements and two angles. The comparison between manual and stereovision measurements showed a high total correlation (r=0.998) and a low variability between operators (SD=0.0004). The average error, lower than 3% and difference in magnitude of error depending by specific traits. In conclusion the stereovision system prototyped could be a good tool to improve phenotyping and enlarge the basis of population involved in breeding programs of horses as well as in other livestock species.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call