一种改进Otsu与人工蜂群优化的图像分割算法  

Image segmentation algorithm based on improved Otsu and artificial bee colony optimization

作　　者：黄翠玲 孔韦韦 呼亚萍 李萌[1,2] HUANG Cuiling;KONG Weiwei;HU Yaping;LI Meng(Xi’an University of Posts and Telecommunications,Xi’an 710121,China;Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing,Xi’an 710121,China)

机构地区：[1]西安邮电大学,陕西西安710121 [2]陕西省网络数据分析与智能处理重点实验室,陕西西安710121

出　　处：《现代电子技术》2021年第11期51-55,共5页Modern Electronics Technique

基　　金：国家自然科学基金面上项目(61772396);陕西省自然科学基金面上项目(2018JM6047)

摘　　要：针对图像分割过程中处理图像计算复杂度高、实时性差等特点,提出一种分割二维Otsu图像的新算法。首先,针对传统二维Otsu算法提出一种新的判别函数;其次,将人工蜂群算法进行优化,反向学习策略产生新蜜源;最后将改进的二维Otsu算法与优化后的人工蜂群算法相结合,通过引领蜂、侦察蜂以及跟随蜂之间的信息共享和分工合作寻找到最佳阈值。仿真实验结果表明,所提算法相较于传统二维Otsu分割算法,能够达到理想的分割效果,且分割速度较快。In view of the high computational complexity and poor real-time performance of image processing in the process of image segmentation,a new algorithm for segmenting two-dimensional Otsu images is proposed.In the algorithm,a new discriminant function is proposed on the basis of the traditional two-dimensional Otsu algorithm first,and then the artificial bee colony(ABC)algorithm is optimized,and the reverse learning strategy is used to generate new honey sources.Finally,the improved two-dimensional Otsu algorithm and the optimized ABC algorithm are combined to find out the optimal threshold value by the information sharing and cooperation among the leading bee,the scout bee and the follower bee.The simulation results show that the proposed algorithm can achieve ideal segmentation effect and faster segmentation speed in comparison with the traditional two-dimensional Otsu segmentation algorithm.

关 键 词：图像分割 人工蜂群算法 Otsu算法改进 判别函数 阈值分割 反向学习策略 

分 类 号：TN911.73-34[电子电信—通信与信息系统] TP391[电子电信—信息与通信工程]