3D Measurement Technology Based On Shape From Silhouette

The 3D measurement of objects of modern testing technology has become an important research field. SFS (Shape From Silhouette) is a passive non-contact 3D measurement method, which is used in 3D modeling and reverse engineering etc. Using SfS method, a simple, low cost, efficient and accurate 3D information acquisition system can be established and the request of the surface material of the object that was measured is low. SFS has also the benefits of non-contact measurement: faster measurements, soft parts, thin workpieces and high-precision workpieces that can not be contacted can be directly measureed. With the continuous development of scientific and technological level, objects in three-dimensional measurement information to scientific research and industrial production occupy a more and more important status.In this paper, the 3D reconsturction technology of object model is the main line. A new method that Shape From Silhouette is applied to measure the objects in coordinate metrology is proposed, that is the vision measurement system is made up of the camera, the coordinate metrology and SFS. The Software system using MATLAB and VC Programming. And with this vision measurement system for measuring three-dimensional objects, through the acquisition images, contour extraction, the application of SFS techniques such as three-dimensional reconstruction of objects in the main steps, complete access to the objects in three-dimensional model.The main features and innovations of this thesis are listed as follows:1. A new method that Shape from Silhouette is applied to measure the objects in coordinate metrology is proposed, that is the vision measurement system is made up of the camera, the coordinate metrology and SfS. Intelligent measurement is carried out on the vision measurement system to 3D reconstruction and consequently the precision is improved. The functions of coordinate metrology is expanded and the applications of coordinate metrology is enlarged by the system.2. Self-designed visual measurement system software. Image acquisition after the software automatically processed by the image processing, extraction of object contours, space carving. Auto-complete reconstruction of three-dimensional model of objects at last.3. Using the automatic thresholding algorithm variance automatic obtain image threshold to extract the contour of the image.4. Mixed use of VC and MATLAB programming.Make full use of the MATLAB in image processing advantages and VC in the interface design advantages.The primary contents in this thesis are listed as follows:1. The mathematical models and the calibration technology of camera are researched.2. Under laboratory conditions, the 3D measurement system was founded based on the technology of SFS.The mathematical model of the system was established and take advantage of the experimental system to measure objects.3. Research on image processing algorithms and accurate data to obtain the outline of objects. The voxel of objects was established, according to the cube in different nodes of the projection point of view of the intersection with the outline of the situation, the use of octree structure refinement level, space carving of the object was realized.4. Preparation procedures,the software system was completed, three-dimensional reconstruction of objects was realized.5,Evaluate the accuracy of the 3D measurement system in order to ensure the accuracy of the measurement results.