A study of road crack detection by the meta-genetic algorithm

This paper describes the detection of road cracks from monochrome images based on an evolutionary model in artificial life. We propose the application of the meta-level genetic algorithm (GA) system to accumulate pseudo-solutions in each generation cycle for the 1st GA. The ability to detect road cracks has been studied on the basis of appropriate template matching controlled by the 2nd-GA which is under the control of the 1st-GA. The number of generation cycles of the 2nd-GA is used as a performance index to evaluate the adaptability of this system.