激光与光电子学进展, 2020, 57 (20): 201102, 网络出版: 2020-10-14
基于卷积神经网络结合改进Harris-SIFT的点云配准方法 下载: 953次
Point Cloud Registration Method Based on Combination of Convolutional Neural Network and Improved Harris-SIFT
成像系统 三维图像采集 点云配准 Harris-SIFT算法 卷积神经网络 深度学习 imaging systems three-dimensional image acquisition point cloud registration Harris-SIFT algorithm convolutional neural network deep learning
摘要
针对传统点云配准方法在处理大型点云模型时存在计算量大、效率低和移动扫描配准实时性较差等问题,提出基于卷积神经网络结合改进Harris-SIFT(Scale Invariant Feature Transform)的点云配准方法。首先改进Harris-SIFT算法,使其可以提取三维空间中点云模型的稳定关键点。进而将关键点的加权邻接矩阵作为卷积神经网络的输入特征图,实现源点云和目标点云关键点的预测匹配。然后基于匹配的关键点,采用迭代最近点(ICP)算法实现点云数据的精配准。相较于传统的点对点配准,所提方法不需要生成对应关系的点描述符,解决全局搜索开销大的问题。实验结果表明,相较于ICP算法,所提方法能够较好地完成即时点云配准,且计算量小,耗时短,效率高。
Abstract
Considering the large amount of calculation, low efficiency, and poor real-time performance of mobile scanning registration when using the traditional point cloud registration method to process large point cloud models, a point cloud registration method based on the convolution neural network combined with the improved Harris-SIFT (Scale Invariant Feature Transform) is proposed. First, the Harris-SIFT algorithm is improved so that it can extract the stable key points of a point cloud model in three-dimensional space. Second, the weighted adjacency matrix of the key points is used as the input feature map for the convolutional neural network. This allows for prediction matching of the key points from the source point cloud and the target point cloud. Then, based on the key points of the matching, the iterative closest point (ICP) algorithm is used to achieve precise registration of the point cloud data. Compared with the traditional point-to-point registration, the proposed method does not need to generate corresponding point descriptors, thus avoiding the problem of a high global search overhead. The experimental results reveal that when compared with the ICP algorithm, the proposed method can better complete real-time point cloud registration, needs minimal calculation, takes a short amount of time, and is highly efficient.
李昌华, 史浩, 李智杰. 基于卷积神经网络结合改进Harris-SIFT的点云配准方法[J]. 激光与光电子学进展, 2020, 57(20): 201102. Changhua Li, Hao Shi, Zhijie Li. Point Cloud Registration Method Based on Combination of Convolutional Neural Network and Improved Harris-SIFT[J]. Laser & Optoelectronics Progress, 2020, 57(20): 201102.