用于室内环境建模的基于形状一致性的立体视觉匹配算法

本文针对室内环境水平特征丰富的特点，提出借助图象变换，使立体视觉图象对中对应水平特征呈现形状一致性，以利于实现特征匹配，并获取物体三维结构的新颖算法，本文地在标定图象及非标定图象两种条件下如何获取所需变换参数进行了深入的理论分析，并对适用于基于形状一致性匹配算法的图象预处理及匹配算法进行了分析与实验研究。从初步获取的实验结果来看，这是一种对室内环境建模较为有效的方法。

A STEREO MATCHING ALGORITHM BASED ON SHAPE IDENTITY FOR INDOOR ENVIRONMENT MODEL BUILDING

A novel stereo matching algorithm has been proposed in this paper based on the observation that there are plenty of horizontal features in indoor environment. In this algorithm, the shape identity of all horizontal feature correspondences is achieved by means of a special image transformation-reprojection transformation, and therefore feature matching can be established based on shape identity with high efficiency. The principle of reprojection transformation using calibrated or uncalibrated images is discussed, and methods of feature extraction and feature matching are suggested. It seems from the preliminary experimental results that the algorithm is an effective and efficient method for indoor environment model building.