 |
The overall aim of this research is to apply the use of
model-based computer vision methodology in order to develop a system for
dentistry to go beyond traditional approaches in diagnosis, treatment
planning, surgical simulation and prosthetic replacements. This computer
vision methodology should obsolete the use and avoid the discomfort of the
mold process used today as well as substantially improves the data
accuracy available to the oral surgeon and dental care personnel. Our
main objectives are as follows: (i) to design a flexible data acquisition
system to capture high quality calibrated intra-oral images of the Jaw,
Teeth, and Gums (ii) to develop methods for accurate 3-D reconstruction
from the acquired sequence of intra-oral images using state-of-the-art
approaches in 3D reconstruction, including shape from shading, and space
carving techniques; (iii) to develop and automate a robust algorithm for
fusion of data acquired with accurate and fast 3D data registration; (iv)
to develop a specific object segmentation and recognition system to
separate and recognize individual 3D tooth information for further
analysis and simulations; (v) to develop algorithms to simulate tooth
movement based on the finite element method and deformable model
approaches; (vi) to develop a database software with a user friendly GUI
interface which stores the acquired digital models in a database that can
be retrieved on demand and shared over computer networks for further
assistance in diagnosis and treatment planning; and
(vii) to enable
the use of haptic feedback interfaces on our 3D jaw models in future work.
 |