Document Type
Thesis
Date of Award
8-31-1990
Degree Name
Master of Science in Electrical Engineering - (M.S.)
Department
Electrical and Computer Engineering
First Advisor
Nirwan Ansari
Second Advisor
Edwin Hou
Third Advisor
Zoran Siveski
Abstract
The problem we consider is to find a subset of points in a pattern that best match to a subset of points in another pattern through a transformation in a least squares sense. Exhaustive search to find the best assignment mapping one set of points to another set is, if the number of points that are to be matched is large, computationally expensive. We develop a genetic algorithm which searches for the best ("almost the best") assignment efficiently. To map the point pattern matching into the framework of a genetic algorithm, we use a fitness function which inversely proportional to the match error, a scheme to encode an assignment between two sets of points into a string, and we also introduce a new genetic operator known as "Mixed-type Partial Matching Crossover." Experimental results have demonstrated the robustness and the fast convergence of the algorithm. The proposed algorithm can be applied to n-dimensional point patterns and any transformation, but we only present results for two-dimensional point patterns and a similarity transformation.
Recommended Citation
Chen, Minghwang, "Point pattern matching by a genetic algroithm" (1990). Theses. 2568.
https://digitalcommons.njit.edu/theses/2568