Inspection of local deformations using 3D scans
|Abstract:||In 3D vision several studies have been conducted to perform the comparison of 3D objects or human models. In fact the comparison of 3D models is sometimes related to the retrieval of models belonging to a large database, and it includes the following procedures: the recognition of objects, their classification in categories or groups according to their types and finally the comparison of the models in the same group in order to identify certain aspects of resemblance or variability in their shapes. The alignment of entire 3D models also enables their comparison for the detection of possible local deformations. There are two sorts of deformation of models: the first mainly affects the position of parts of the object or the whole object in cases such as the articulation, the stretching or the contraction of parts without altering the volume of the object. The second type of deformation results from a change in volume due to mass variation. Other more complicated cases occur when the change in form includes both kinds of deformation at the same time. In this project we propose an “inspection method” based on an exact registration technique of the rigid parts of the deformable models that have undergone local deformations. When the deformation is local, it means that there is a change of some parts of the model rather than a global deformation of the whole object. We propose an approach that is able to detect the region of deformation. This approach exploits two kinds of local deformations: non-volume conservative transformations caused by either the inflation or the deflation of models’ parts and volume conservative transformations caused by bending, twisting, rotating or displacement of either internal or external parts of the model. In general, the internal parts are the middle parts of an object, while the external parts are the parts at the extremities or the end parts. We have experimented on different cases of deformations on man-made objects and mechanical parts for industrial inspection of possible artifacts and defects, and for quality control purposes as well. We have also applied our comparison method on parts of the human body. Examples relevant to medical applications are the detection and inspection of a lesion which are also tested in this work. Our algorithm is not only successful for detecting local change but it also achieves high accuracy comparable to other well-known industrial inspection technique.|
|Document Type:||Thèse de doctorat|
|Open Access Date:||23 April 2018|
|Collection:||Thèses et mémoires|
All documents in CorpusUL are protected by Copyright Act of Canada.