非特征点双目测距技术研究

提出了一种局部稠密匹配与人工干预相结合的测距方案，利用非特征点与特征点的位置关系，构建“最小矩形”以缩小匹配范围，再应用NCC（归一化互相关）算法对非特征点进行稠密立体匹配，最后根据双目测距原理直接获取非特征点的距离信息。该方法能任意选取图像上的非特征点进行实时距离测量，具有精度高、速度快和操作性强等优点。

Research on distance measurement technology of non-feature points binocular stereo vision

This paper proposes a non-feature point distance measurement combined with local dense matching feature points and human intervention. The method build a ＂minimum rectangle ＂ to narrow down the matching range by using the position rela- tionship between the non-feature points and the feature points. Then NCC algorithm is applied to non-feature points＇ dense stereo matching. Finally, according to the principle of binocular rangefindcr features direct access to non-point distance information. Exper- imental results prove this method with high accuracy , fast speed and maneuverability.