Mesh Segmentation - A Comparative Study
SMI '06 Proceedings of the IEEE International Conference on Shape Modeling and Applications 2006
Object Pose Detection in Range Scan Data
CVPR '06 Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2
Face recognition with irregular region spin images
SCIA'07 Proceedings of the 15th Scandinavian conference on Image analysis
| Hi-index | 0.00 |
In current design and manufacturing processes, the verification of tolerances is mainly focused on dimensional tolerances. As a general rule, manufacturing instructions specify dimensional tolerances and surface qualities. However, geometric tolerances are hardly ever specified, and this often leads to a lack of adequacy of the design. A better adequacy of this design would certainly produce a significant reduction in manufacturing costs. The lack of trained engineers, as well as the non-availability of the appropriate software, has relegated geometric tolerances, at the very best, to a second place. One of the greatest difficulties when using the current software solutions is the partition and identification of the meshes obtained from a scanning process on specific features. These processes can hardly be carried out via automation because they are subject to the accuracy of the tool and the skill of the technician in charge of processing the meshes. This paper presents a semi-automatic process to detect non-ideal, cylindrical features in point clouds. Its aim is to identify and extract the points from these features in order to implement them into the verification algorithms of possible associated geometric tolerances. In this process, spin-images, usually developed in the detection of shapes, are used as a graphic tool.