Performance Evaluation of Evolutionary Algorithms for Edge Detection

C.M. Ng, C.K. Ckan, and F. Chun (PRC)

Keywords

Genetic Algorithm, Tabu Search, Evolutionary Tabu Search, Chromosome, Crossover, Mutation.

Abstract

One of the major problems with most of the existing operators in edge detection is huge search space, with the advent of high quality image capture devices, the resolution of the captured becomes bigger and bigger. Therefore, without optimization, the task for edge detection in image processing is time consuming and memory exhausting. This paper presents an experiment which evaluate the performances of three different evolutionary algorithms on edge detection. The algorithms are Genetic Algorithm (GA), Tabu Search (TS) and, Evolutionary Tabu Search Algorithm (ETS). Experiment results shown that ETS generates the best edge detection result with the longest processing time. TS yields the second best searching result with the shortest processing time, and GA products the worst edge detection result with processing time 5 times longer than TS but about 10 times faster than ETS.

Important Links:



Go Back