基于迭代最少点和遗传算法的点云粗配准算法
Point Cloud Coarse Registration Algorithm Based on Iterative Minimum Point and Genetic Algorithm

为了兼顾点云配准过程的时间和精度，提出了基于迭代最少点和遗传算法的点云粗配准算法。将源点云和目标点云的总点云进行下采样，以采样后总点云的数量作为遗传算法的目标函数，采用遗传算子指导解的搜索方向，通过新种群的迭代使下采样总点云数量最少，快速得到点云粗配准的结果。通过对6组不同的点云，以及采用多种算法进行对比试验，结果表明，该算法在保证配准精度的同时，耗时稳定在24 s左右，且对待配准点云模型无特殊要求，鲁棒性较强。
To balance the time and precision of the point cloud registration process, a point cloud coarse registration algorithm based on the iterative minimum point and genetic algorithm is proposed. The total point clouds of the source point clouds and target point clouds are down sampled, with the number of the total point clouds after sampled as the target function of the genetic algorithm, the genetic operator guides the search direction of the solution, the total number of sampled point clouds minimizes through the iteration of the new population, and the results of point cloud coarse registration are obtained quickly. By comparing the tests of six different point clouds and using various algorithms, the results show that the proposed algorithm takes about 24 seconds stably while ensuring the registration accuracy, and treats the registration point cloud model without special requirements and has strong robustness.