Abstract
A simple, fast and accurate technique for direct detection of 3-D shape of rotated objects based on light line projection is presented. This technique is proved by simulation and experimentally verified. To obtain the surface shape, the object is rotated at an angle with an electromechanical device. In each rotation, the object is scanned by a light line. To determine the surface data, disturbs of the light line are measured on a reference point and on the object surface. This line displacement is measured by Gaussian approximation, which corresponds to the intensity model of a light line. In this technique, the rotation angle and origin are calculated by recognition of a light line pattern. Recognition method uses Hu Invariant Moments to detect light line pattern. A set of images is processed to obtain the complete 3-D shape. Also, the necessary time for processing a light line by using this method is presented. This echnique is tested with real objects and its experimental results are presented.
