XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE Applying 3D-Scan Systems RangeVision for Precision Preparing of Polygonal 3D-Models Mikhail A. Petrov Material forming and additive technologies FSBEI HE “Moscow Polytechnic University” Moscow, Russia m.a.petrov@mospolytech.ru Ibrahim S.A. El-Deeb Material forming and additive technologies FSBEI HE “Moscow Polytechnic University” Moscow, Russia Production Engineering and Mechanical Design Tanta University Tanta, Egypt ibrahim.eldeeb@f-eng.tanta.edu.eg Abstract—Many different applications of product inspections have found a significant advantage by the use of 3D-scanners, especially when working with complex surfaces, where traditional inspection tools have significant limitations. This work deals with the structured light 3D-scanning based on fringe-pattern and obtaining polygonal mesh models (stl-models) of several end- and half-products, manufactured by different operations of blank production. In the study, the preparing of point clouds are discussed as well. 3D-scanning is possible to apply for getting the high resolution and accurate results for both metallic and non-metallic materials. Careful 3D-model reconstruction provides the size tolerances up to 0.05 mm for the rough surfaces after stamping and casting operations and up to 0.60 mm for wavy surfaces after 3D-printing by FFF-method. The automatic repairing and healing of the 3D-models can cause artefact building and trespassing the accepted thresholds for tolerances. Keywords—machine vision, 3D optical scanning, light fringe pattern, polygonal 3D-model, detalization, point cloud I. INTRODUCTION In mechanical engineering of modern production, the reproducibility of results and quality of products are highly valued, which is approved by the specification of the customer and fixed by process documentation. With this approach, the control of the product production is carried out after each production step (e.g. post-operative control, intermediate control, selective final control etc.). To control quality of product according to the drawing dimensions a contactless method using optical/laser system (3D-scanner, tracker, LIDAR), which is operating in either online/inline mode or post-mode, is used. The dimension evaluation is carried out by dimension deviation field or field of tolerance. Thus, two tasks are solved at once: automatic dimension and contour control, and saving the measurement results in the product database (DB), which stores the step-by-step history of the product manufacturing. During the last decade the amount of such investigations were carried out elsewhere [1–15] and represent the applications of machine vision for forging/stamping, casting and 3D-printing facilities of blank production. A. 3D-scanning Conventionally, 3D-scanners are divided into optical and laser. The first type operates with the structured light (normally LED) with fringe-pattern (SFP). The second type operates with the laser-based light sources. The working scheme of the first type of 3D-scanner device could be found elsewhere [16]. Working on the triangulation principle, as well as simple laser scanners, and in contrast to the laser scanner, it projects the black-and-white fringed patterns (lines, dots, asterisks etc.) on the object. The image sensor determines the height of the object across the entire projection plane of the pattern, consisting, for example, of a set of lines, and not along one line, as in a laser linear scanner. Further, the reflected light falls on the sensitive CMOS stereo cameras (monochrome or colour), the coordinate of the points in space is determined. After obtaining a set of points or a point cloud (PC), it is necessary to cross-link them using polygon elements (PGE), thus creating a polygon mesh (PGM). In the future, a texture is applied to the PGM for the reconstruction of a photorealistic PGM-model or stl-model. B. Detalization Some of the areas of the object have to be controlled during manufacturing. The control stage is referred either to the collecting the information about the detail or to the defects, which occur during its production. In the Fig. 1 the defect like crack on the body of the 3D- printed connecting rod was detected with the help of smartphone camera of 13.0 MPix (without macro mode, a), USB-microscope (magnification up to 300x, b) and digital camera of 12.1 MPix (with macro mode). The cracks are good recognizable but the detalization differs from each other. From the basic image (Fig. 1a) there is no possibility to determine the contour and geometrical sizes of the crack. To improve the quality the microscope capturing (Fig. 1b), the photogrammetry extended with macro mode (Fig. 1c) or its combination (not presented) can be applied. In the case of the 3D-scanning the similar result has been obtained for the middle calibration field (CF).