Plane extraction method, system, device and storage medium based on point cloud data

Abstract

A plane extraction method based on point cloud data is disclosed. A data grid and a predetermined scan line are recover from that point cloud data, and based on this, another pseudo-scan line is constructed, and then the unnecessary points in the scanning line are removed by the line simplification algorithm. The original scanning line is divided into each scanning line segment based on the characteristic points, and the main direction of each scanning line segment is used as the reference for clustering processing to obtain each plane fragment composed of line features, and finally, the planefragments with the same or similar normal direction and overlapping regions are combined to obtain the target plane. By the way of line feature, the existing algorithm can get rid of the dependence on the high point cloud density, and at the same time, it can significantly reduce the number of normal contrast, and improve the efficiency. At that same time, the invention also discloses a region ofinterest determination device, a device and a computer-readable storage medium, which have the above-mentioned beneficial effect. 本申请公开了一种基于点云数据的平面提取方法，从点云数据中恢复得到数据格网和既定扫描线，并基于此构建得到另外的伪扫描线，之后再利用线简化算法去除扫描线中不必要的点，得到保留下来的特征点，并基于特征点将原扫描线拆分为各扫描线段，以各扫描线段的主方向为参考依据进行聚类处理，得到由线特征组成的各平面碎片，最后将拥有相同或相似法向且存在重叠区域的平面碎片进行合并得到目标平面。通过以线特征的方式摆脱了现有算法对高点云密度的依赖，同时也能够显著减少法向对比的次数，提升了效率。本申请还同时公开了一种感兴趣区域确定装置、设备及计算机可读存储介质，具有上述有益效果。