Asynchronous demodulation technique for use in vision sensor image classification and segmentation

摘要: This paper presents a novel technique for classification and segmentation of multiple-object image scene. Each object in the scene is tagged by flashing LED operating at specific frequency. The vision sensor, based on this technique, demodulates captured light signal, omits background illumination, performs assigning unique ID-tag to each region its process performed in-pixel an asynchronous demodulation frequency identification circuit, which designed standard 0.6µm CMOS technology. Simulation results confirm validity proposed structure. At frame rate 250 fps power consumption 2.6µW/pixel, relatively low compared with those sensors similar functionality. structure intended use as power, inexpensive solution robot visual position feedback localization. Copyright © 2009 John Wiley & Sons, Ltd.