Robotics Research Technical Report: Shape Description With a Space Variant Sensor: Algorithms for Scan-Path, Fusion and Convergence, Over Multiple Scans (Classic Reprint) - Softcover

Yehezkel Yeshurun

 
9781332093748: Robotics Research Technical Report: Shape Description With a Space Variant Sensor: Algorithms for Scan-Path, Fusion and Convergence, Over Multiple Scans (Classic Reprint)

Synopsis

This book explores the potential of space variant sensors in machine and robotic vision, inspired by the strongly space variant nature of the human visual system. The author, an expert in computational neuroscience, utilizes a complex logarithmic mapping function to accurately model the human visual representation and provides realistic estimates of the nature of such a unique imaging system. The author delves into three primary algorithms related to the blending of a single scene from multiple frames: fusing space variant contour-based scenes with different fixation points into a single multi-scan view; an attentional algorithm that recursively examines the current knowledge of the scene to optimally choose the next fixation point based on focusing attention in regions of maximum boundary curvature; and a simple metric for evaluating convergence over scan path. Through detailed analysis and illustrative examples, this book sheds light on essential concepts such as image data compression, visual attention, and convergence, highlighting the potential of space variant imaging systems in both machine and biological vision.

"synopsis" may belong to another edition of this title.

Other Popular Editions of the Same Title