论文部分内容阅读
针对目前基本遗传算法在优化图像分割算法中存在的易于早熟、陷入局部最优的不足,以最大类间方差函数为适应度函数,提出了一种基于改进遗传算法的图像阈值分割算法。对交叉、变异算子进行自适应改进,同时将模拟退火算法融入到遗传算法中,使得对个体的评价更合理,既能克服种群退化现象,又改善算法的全局搜索能力,避免遗传算法陷入局部最优。实验结果显示,与Otsu图像分割法以及基于遗传算法的图像分割方法相比,使用该方法得出的阈值范围更加稳定,执行效率更高,在图像分割中获得的分割效果更佳。
Aiming at the shortcomings of the current basic genetic algorithms, which are easy to premature and fall into the local optimality, and the maximum between-class variance function is the fitness function, an improved image segmentation algorithm based on the improved genetic algorithm is proposed. The crossover operator and the mutation operator are improved adaptively. At the same time, the simulated annealing algorithm is incorporated into the genetic algorithm, which makes the evaluation of the individual more reasonable. It can not only overcome the population degeneracy but also improve the global search ability of the algorithm, Optimal. The experimental results show that compared with Otsu image segmentation and genetic algorithm based image segmentation, the threshold range obtained by this method is more stable and the execution efficiency is higher, and the segmentation result obtained in image segmentation is better.