Roger Dufour

Target Classification and Parameter Estimation with a Deformable Template Library

Date: Thursday April 24, 2003
Time: 1:00pm
Room: 302 Stearns

Abstract:

Two common image processing problems are determining the location of an object using a template when the size and rotation of the true target are unknowns and classifying an object into one of a library of objects again using a template-based matching technique. When employing a maximum likelihood approach to these problems, complications occur due to local maxima on the likelihood surface. In this thesis, we have demonstrated a technique for object localization that employs a library of templates ranging from a smooth approximation template to the exact template with varying degrees of detail. Successively estimating the geometric parameters (i.e. size and rotation) using these templates achieves the accuracy of the exact template while remaining within a well-behaved ``bowl'' in the search space which allows standard maximization techniques to be used. Further, this technique is extensible to solve the classification problem using a multiple template library. We introduce a steering parameter that at every scale, allows us to compute a template as a linear combination of templates in the library. The algorithm begins the template matching using a smooth blob which is the smooth approximation common to all templates in the library. As the location and geometric parameter estimates are improved and detail is added, the smooth template is ``steered'' towards the most likely template in the library and thus classification is achieved. In this thesis, we have developed these algorithms and demonstrated their performance against both simulated data and real data.

Thesis Committee:
Prof. Eric Miller, Northeastern University (Advisor)
Prof. Dana Brooks,Northeastern University
Prof. David Castañon, Boston University
Prof. Nikolas Galatsanos, University of Ioannina
Dr. Michael McCormack, Textron