【摘要】 针对不同坐标系下部分重叠的稀疏坐标点集，提出一种基于距离哈希的同名点快速稳健匹配算法。将各点与其邻近点的距离关系映射成一个二进制码身份标签，通过身份标签相似度计算，找出两个点集中满足设定阈值的候选匹配点对，从而建立初始匹配关系。据此计算刚体变换矩阵对两组点集进行配准，确定两组点集之间的精确匹配关系。实验结果表明：该算法不仅速度快、准确率高、对于噪点和低重叠度具有稳健性，而且对两个点集之间的初始相对位置没有任何限制。更多还原

【Abstract】 Based on distance hash, proposes a fast and robust matching algorithm with homonymous points for partially overlapping sparse coordinate point tsets in different coordinate systems. A binary code identity tag is mapped according to the distance relationship between each point and its adjacent points. Through the similarity calculation of identity tags, the corresponding similar point pairs meeting the set threshold in two point sets are found to establish the initial matching, based on which, the rigid body transformation matrix is calculated to register the two point sets, and the precise matching between the two point sets is defined. The experiment results show that the proposed algorithm is fast, accurate, robust to noises and low overlapping, and hasno restriction on the initial relative position between two point sets.更多还原