INTRODUCTION TO ACTIVE VISION
Real-time vision applications can achieve their required performances only by applying the appropriate computations on the relevant data, and at the right time. A ‘smart’ selection of data, located almost anywhere in the wide field of view, is generally performed at the early stages of the analysis, to broadly detect what is present in the scene, and where it is. The human camera-eye maneuvers for position and ‘localizes’ the regions of interest thanks to a sophisticated control. The highest concentration of sensing transducers occurs in the fovea, which occupies just a small portion of the field of view: about 120 minutes of arc around the visual axis; meanwhile, the sensing field is extended broadly, over 180 ...
Get Artificial Vision now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.