M. Hirose, T. Miyasaka, and K. Araki (Japan)
Integration, Registration, Reconstruction, Serial Range Image, Extended Gaussian Image, Geodesic Dome
In this paper, we present a method for automatic integration of serial range images measured under the condition free from any artificial constraint. As first step, serial range images are captured by real-time 3-D measurement system developed in our laboratory. Then, four intersection points between four real and/or virtual planes extracted from Extended Gaussian Image (EGI) of the target are estimated in each range image and four corresponding points between two serial range images are determined from thus estimated intersection points and EGI, automatically. Next, affine transformation matrices, which give mutual position relationship between two serial range images, are derived in order from thus determined four corresponding points. Concatenating thus derived affine transformation matrices in order, registration of all serial range images are performed by transforming them to one coordinate system of arbitrary chosen range images. Finally, thus registered serial range images are reconstructed by means of geodesic dome type data structure, sampling regions of which are isotropic and so it may faithfully represent 3-D shape data. The validity of our integration method by registration and reconstruction was proved by experiments on processing time and accuracy.
Important Links:
Go Back
