An evolutionary tabu search approach to optimal structuring element extraction for mst-Based shapes description

Abstract

Optimal structure element extraction is a key step in the application of mathematical morphology to various image processing tasks and shape description problem in computer vision. In this paper, we propose a novel optimization technique called evolutionary tabu search (ETS) to solve optimal structure element extraction problem for MST-based shape description. Specifically, we incorporates “the survival of strongest” idea of evolution algorithm into tabu search. This new method has the ability to find the global optimum, which not only keeps the advantages of tabu search and Genetic Algorithms, but also overcomes some of their shortages. Specifically, by comparing our algorithm with the existing other global optimization methods (such as genetic algorithm, Simulated annealing and tabu search), we find that the ETS is more practical and effective, which also yields good near-optimal solutions and has better convergence speed

Keywords:

• Shape Description
• Mathematical Morphology
• Shape Matching
• Optimal Structure Element
• Simulated Annealing Algorithm
• Genetic Algorithm
• Tabu Search
• Model-Based Vision