改进蜜獾算法优化OTSU的图像分割研究

Image Segmentation on Improved HBA to Optimize OTSU

群体智能算法结合图像分割技术已经成为图像处理领域中的新热点,传统的图像分割方法需要大量的人力和时间,蜜獾算法(honey badger algorithm,HBA)可以通过模拟蜜獾觅食的行为来执行优化任务,在寻找解决问题的过程中可以逐步逼近最优解来实现图像分割任务;通过反向学习策略改进蜜獾种群的初始化,提高种群多样性和分布平衡,从而提高算法的整体搜索能力;引入柯西变异因子,对算法计算得到的可行解进行扰动,使算法更易于跳出局部最优,增强算法的局部搜索能力和收敛精度;选取三幅测试图像进行分割验证,实验结果显示,融合改进蜜獾算法和二维OTSU算法得到的分割图像精度更高、效果更细致,验证了方法的有效性;综上所述,改进蜜獾算法具有更好的鲁棒性和泛化性,优化的二维OTSU算法可以更好地处理复杂场景和图像。

Population intelligence algorithm combined with image segmentation technology has become a new hot spot in the field of image processing.Traditional image segmentation methods require a lot of manpower and time,honey badger algorithm(HBA)can perform the optimization task by simulating the behavior of honey badger foraging,and can gradually approach the optimal solution to achieve the image segmentation task in the process of finding a solution to the problem;The initialization of the honey badger population is improved by the backward learning strategy,which improves the population diversity and distribution balance,thus enhancing the overall search ability of the algorithm;Cauchy variation factor is introduced to perturb the feasible solutions calculated by the algorithm,which makes the algorithm easier to jump out of the local optimum and enhances the local search ability and convergence accuracy of the algorithm;three test images are selected for segmentation verification,and the experimental results show that the segmented images obtained by fusing the improved HBA and 2D OTSU algorithm are more accurate and detailed,which verifies the effectiveness of the method.In summary,the improved HBA has better robustness and generalization,and the optimized 2D OTSU algorithm can better handle complex scenes and images.