Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Lane Department of Computer Science and Electrical Engineering

Committee Chair

Hany Ammar.


The Automated Dental Identification System (ADIS) is a Post Mortem Dental Identification System. This thesis presents dental image registration, required for the preprocessing steps of the image comparison component of ADIS. We proposed a multi resolution dental image registration based on genetic algorithms. The main objective of this research is to develop techniques for registration of extracted subject regions of interest with corresponding reference regions of interest.;We investigated and implemented registration using two multi resolution techniques namely image sub sampling and wavelet decomposition. Multi resolution techniques help in the reduction of search data since initial registration is carried at lower levels and results are updated as the levels of resolutions increase. We adopted edges as image features that needed to be aligned. Affine transformations were selected to transform the subject dental region of interest to achieve better alignment with the reference region of interest. These transformations are known to capture complex image distortions. The similarity between subject and reference image has been computed using Oriented Hausdorff Similarity measure that is robust to severe noise and image degradations. A genetic algorithm was adopted to search for the best transformation parameters that give maximum similarity score.;Testing results show that the developed registration algorithm yielded reasonable results in accuracy for dental test cases that contained slight misalignments. The relative percentage errors between the known and estimated transformation parameters were less than 20% with a termination criterion of a ten minute time limit. Further research is needed for dental cases that contain high degree of misalignment, noise and distortions.