Notes on Stochastic Processes on Manifolds
Roger W. Brockett · Birkhäuser Boston eBooks · 1997
Vision, as a sensing modality, differs from sensing a position of a shaft or the voltage from a thermocouple in that the data comes in the form of a two dimensional array coded in such a way that the location of objects, typically the information to be used in defining the feedback signal, must be extracted from the array through some auxiliary process involving image segmentation, computation of centroids, etc. It is both a blessing and a curse that sensing via light is a noncontact, “remote” modality so that the meaning of what is being sensed depends in a significant way, on the positioning of the vision sensor relative to the objects being sensed. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.