دانلود رایگان مقاله لاتین ثبت ابر نقطه از سایت الزویر
عنوان فارسی مقاله:
ثبت خودکار ابر نقطه از چند دیدگاه با استفاده از اهداف حوزه
عنوان انگلیسی مقاله:
Automated registration of multi-view point clouds using sphere targets
سال انتشار : 2015
برای دانلود رایگان مقاله ثبت ابر نقطه اینجا کلیک نمایید.
مقدمه انگلیسی مقاله:
1. Introduction
Measuring the 3D shape of an object is frequently considered in recent manufacturing processes to maintain the quality of products and to control the fabrication process. It is more important in fabrication processes when multiple parts are assembled. The actual 3D geometric shape of each part can be compared with its CAD (Computer Aided Design) model in advance to identify any potential problem in the assembly and correct it if necessary [1]. In the shipbuilding industry, such demand is growing. A ship or an offshore structure is a custom-ordered product, and few standardized parts are used to build a ship. This means that every part should be uniquely designed and fabricated, and the accurate assembly of the parts is critical for the quality of the final product. To achieve the desired quality, the actual shapes of the fabricated parts need to be evaluated against their CAD models to verify that the models have been made accurately. Here, an efficient method to acquire the 3D shape is to use a laser-based 3D scanner. A 3D scanner is a line-of-sight device. Moreover, the field of view that the scanner covers is limited. Therefore, multiple scans in different positions and directions are typically required to scan a large object. The acquired point clouds must be merged into one data set for correct representation of the shape of the object. The merged point clouds, however, fail to represent the shape of the object because each point cloud is defined with respect to a different coordinate system, as illustrated in Fig. 1. Fig. 1(a) and (b) show two point clouds measured from two different locations and directions. Merging the measured point clouds naively with respect to one reference coordinate system produces misaligned point sets as illustrated in Fig. 1(c). The circled parts in Fig. 1(a) and (b), which should be the same object, are not aligned and instead produce two separate objects. Therefore, aligning point sets with respect to a reference coordinate system is necessary to obtain a set of points that correctly represent the shape of a target object; this process is called registration. Registration is a problem that has attracted the attention of researchers from around the world, and diverse algorithms for the problem have been proposed in the related literature. With the significant growth of data in size and complexity due to the extensive use of high-performance scanners, it is difficult for a human to handle them efficiently for registration. Moreover, the linear extensions of the existing algorithms often fail to handle such data. Therefore, it is necessary to develop a method for registration that can handle data of large size robustly and efficiently and that can run automatically. In this paper, the problem of registration of point clouds is addressed, and a novel method for automatic registration of two point clouds using sphere targets is proposed. The overall process for registration consists of filtering, sphere detection, initial registration and fine registration. The point clouds, each of which contains sphere targets, are produced by a 3D scanner. They are
کلمات کلیدی:
High-Precision Registration of Point Clouds Based on Sphere Feature ... www.mdpi.com/1424-8220/17/1/72/pdf by J Huang - 2016 - Cited by 1 - Related articles Dec 30, 2016 - Abstract: Point cloud registration is a key process in multi-view 3D .... So some sphere targets with specific features are introduced and fixed ... [PDF]Automated registration of multi-view point clouds using sphere targets iranarze.ir/wp-content/uploads/2016/09/E73.pdf Oct 21, 2015 - This paper addresses the problem of automated registration of multi-view point clouds generated by a 3D scanner using sphere targets. First ... Automated registration of multi-view point clouds using sphere targets ... https://www.semanticscholar.org/...multi-view-point-clouds.../768380472051251845e0c... Semantic Scholar extracted view of "Automated registration of multi-view point clouds using sphere targets" by Dongho Yun et al. [PDF]Conference paper - ISPRS-Annals www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/.../isprsannals-I-3-173-20... by PW Theiler - Cited by 33 - Related articles TLS point clouds without the need for artificial targets. 1. ... objects, which are invariant to the scanner viewpoint (e.g. planes, spheres, cones). Often, they are .... with RANSAC, embedded in a multi-scale pyramid, which on the one hand ... [PDF]Registration of 3D Point Clouds for Ship Block Measurement - SNAME www.sname.org/HigherLogic/System/DownloadDocumentFile.ashx?...f5cd182d... Moreover, a registration method for multiple point clouds is presented, which ... for registering point clouds using sphere targets. .... interaction in a view window. [PDF]automatic registration of terrestrial laser scanner point clouds using ... https://www.ethz.ch/content/dam/ethz/special-interest/baug/igp/.../theiler2012.pdf by PW Theiler - Cited by 33 - Related articles TLS point clouds without the need for artificial targets. 1. ... objects, which are invariant to the scanner viewpoint (e.g. planes, spheres ... registration with reflective targets. A related approach finds non-reflective spheres (Franaszek et al., 2009). ... constraints between simple configurations of multiple tie points. He et al. [PDF]Fast and Robust Multi-View 3D Object Recognition in Point Clouds graphics.usc.edu/cgit/publications/papers/3dv2015.pdf by G Pang - Cited by 2 - Related articles 3D point cloud into several 2D depth images from multiple viewpoints ... for 3D object recognition is based on multi-view projection, ... Shang and Greenspan [22] use view sphere ... data scene and the search target object are projected for 2D. An Improved Algorithm of Precise Point Cloud Registration :: Science ... article.sciencepublishinggroup.com/html/10.11648.j.pamj.s.2015040501.19.html Secondly, based on a normal-based algorithm of registration sphere center, the points ... The main multi-view point cloud registration methods are ICP algorithm and ... from a certain point to its closest point, and takes it as the matching target ...
