Maximum Variance Combined with Adaptive Genetic Algorithm for Infrared Image Segmentation
Huixuan Fu, Yuchao Wang, Liangliang Han · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2015
Maximum Variance Image Segmentation method (Otsu) is a popular non-parametric method in image segmentation.However, it is large amount of computation and poor real-time quality have limited its further application.To solve these problems, a new approach based on an adaptive genetic algorithm (AGA) and Otsu are proposed, which using between-class variance as fitness function, automatically adjusts the optimal threshold.The adaptive genetic algorithm selects crossover probability and mutation probability according to the fitness values, reduces the convergence time and improves the precision of genetic algorithm, insuring the accuracy of parameter selection.The experimental results show that the proposed method is better than the original Otsu, the AGA-Otsu can provide better effectiveness on experiments of infrared image segmentation, decrease processing time.