3 D Reconstruction Based On Image Key Technology And Application

We need to model the3D curve and geometry structures of the products in industrial production. This thesis deals with a method of3D reconstruction from images which are obtained by only one hand-held camera on the basis of computer vision and computer graphics.The main contents of this thesis include:1) the method of processing digital images;2) the extracting skeleton of the characteristics curves extraction;3)3D curves reconstruction based on correspondences among multiple digital images.A method based on mathematical morphology is proposed to extract the skeleton of the characteristics curve in this paper. LOG operator can only detect the edge rather than the central axis of the characteristics curve. So it can not be used to reconstruct the curve of three-dimensional model. We propose a method based on mathematical morphology which is used to extract the skeleton of the characteristics curve.We complete the curve reconstruction according to the following steps. Images of the shoe-last were acquired by the calibrated camera, and skeletons of the characteristics curve of shoe-last are extracted. After obtaining the camera parameters and poses, the active contour model is used to extract the image curves. The correspondence of the images curves is estimated based on the theory of the epipolar geometry. Then a3D curve reconstruction algorithm is studied and put forward. The digital model of the shoe-last is reconstructed after developing the basis of above research. Experimental results show that the reconstructed model of the shoe-last using the method has higher precision and the equipments reconstructing3D model is simpler and cheaper.