中南大学学报（自然科学版）2016，Vol.47Issue(7)：2304-2310,7.DOI:10.11817/j.issn.1672-7207.2016.07.018
基于模糊逻辑的多尺度小基高比立体匹配方法
Multiscale stereo matching method for small baseline based on fuzzy logic
摘要
Abstract
A stereo matching method for small baseline was proposed to solve radiometric variations and depth accuracy in small baseline stereovision. Firstly, the multiscale space was constructed using wavelet transform. The raw matching cost was calculated by fuzzy logic based similarity function and cross based window. Then the“winner takes all”strategy was used to compute integral disparities. Finally, the sub-pixel stereo matching method based on iterative phase correlation was employed to obtain sub-pixel disparities on the basis of integral disparities. The results show that the proposed stereo matching method overcomes the radiometric variations problem and achieves the dense disparity map with sub-pixel precision higher than 1/20.关键词
立体匹配/小基高比/自适应窗口/模糊逻辑/亚像素视差Key words
stereo matching/small baseline/adaptive window/fuzzy logic/sub-pixel disparity分类
信息技术与安全科学引用本文复制引用
马宁,门宇博,李香,门朝光..基于模糊逻辑的多尺度小基高比立体匹配方法[J].中南大学学报（自然科学版）,2016,47(7):2304-2310,7.基金项目
国家自然科学基金资助项目(61100004，11547157)；黑龙江省自然科学基金资助项目(F201320)(Projects(61100004,11547157) supported by the National Natural Science Foundation of China （61100004，11547157）
Project(F201320) supported by the Natural Science Foundation of Heilongjiang Province) （F201320）
