Generation of dense range maps by data fusion from active andpassive colour stereo vision

摘要

The problem of correspondence search in a stereo system may bealleviated if an active light source projects colour codes onto thescene. To reduce the amount of ambiguous or false matches, additionalinformation drawn from feature-based passive stereo techniques can beused In this paper we discuss the application of fuzzy data fusiontechniques to a dual-mode stereo system. Information from passive stereois used whenever the active procedure cannot find an unambiguous match.The algorithm presented in the paper uses a fuzzy rule-based approach tointegrate crisp as well as fuzzy input data. It is shown that thecombination of data obtained in both modes of operation may lead to asignificant increase in the number of matches meeting a certaincredibility criterion. We present some of our experimental results basedon real stereo image data and conclude the paper with a discussion ofperspectives for possible future research