Vision System for Robotic Attacher

摘要

In certain embodiments, a system includes a controller operable to access a first image generated by a first camera. The controller determines a reference point from at least one main feature of a dairy livestock included in the first image. The controller is further operable to access a second image generated by the second camera. The second image includes at least a portion of an udder of the dairy livestock. The controller determines a location of a teat of the dairy livestock based on the second image.