Search

Terrestrial laser scanning this time became one of the most important tool which delivers reliable and rapid data, describes 3D shape and sometimes colours of analyzed objects. Robustness of this measurement approach depends on properly performed design stage in this data acquiring technology. There are many issues during data acquisition, which complicate the process. Sources of these problems are usually: reflectiveness of registered objects, angle between laser ray and a normal vector to the surface and the behaviour of the laser light on the edge of the object. Environmental limitations such as dust and fog in the air makes also difficulties with proper registrations of three-dimensional form. This paper presents research results of testing the impacts of above mentioned factors on acquisition of laser scanning data. The main tasks consisted of the following subjects: -- differentiated conception of control points displacement, -- analysis of reflection angle, intensity and its influence on point registration, -- research into accuracy of scans orientation on the base of signalized control points Results of described tasks ought to improve accuracy and especially repeatability of acquired objects surface from laser scanning technology in close range. [ABSTRACT FROM AUTHOR]

The subject of this paper is a preliminary analysis of the algorithms for automatic spatial orientation of the clouds of points obtained by using a terrestrial scanner. In the proposed data processing method, the clouds of points were treated as images projected on a spherical plane of reference, additionally enriched with a depth map. The algorithms and a library of the sets of functions used in CV were presented. The implementation of a free C++ language library with algorithms used in CV was analysed. The correctness of the tie points search and operation time was examined. Clouds of points obtained from a terrestrial laser scanner Z+F 5006H were used as the source data. The analysis presented the algorithms linking the photogrammetric approach with the approach supported by CV. [ABSTRACT FROM AUTHOR]

Projection of appearance and geometry of complicated industrial objects is a complex task, which requires that appropriately scheduled and prepared measurements are performed. Such objects must be accurately inventoried, but their complicated shapes often makes the access to such objects and visibility of their entire surface very difficult. Prepared documentation was often developed in the form of sketches, plans or maps, which were additionally amended by photographic documentation. Visualisation of objects in such a form is often insufficient. Due to the increasing number of industrial objects, as well as technological development, the need to modify methods of inventorying has appeared. Designing of new elements with the use of CAD/CAM tools requires that an object is described in three dimensions. Measurements were initially performed with the use of tacheometers, allowing for obtaining spatial co-ordinates of each point. Such technique required the utilisation of a mirror and involvement of the second person. Zones which are excluded from the possibility of visits by humans often exist within the areas of industrial objects. That is why mirror-less total stations appeared in the market, which allow for contact free measurements. However, measurements lasted too long and they were not cost effective. Basing on this technology a laser scanner was constructed. It automatically measures millions of points [5] and operates much faster than a total station. The terrestrial laser scanning technology produces a three-dimensional description of geometry of scanned objects. The accuracy of measurements performed by the scanner does not differ from the accuracy of measurements performed by precise tacheometers and the speed of measurements reaches about o million of points per second. Integration of laser scanning and digital images allows for creation of a realistic 3D model. Data acquired by means of the laser scanner may serve for inventorying complex industrial objects and for valuable analyses, which could not be performed using conventional techniques. Operating the "cloud of points" includes realistic visualisation and precise measurements in a space transformed to the virtual reality and it is the basis for advanced spatial modelling and reverse engineering [2]. [ABSTRACT FROM AUTHOR]