摘要
针对现有图像分割算法中计算复杂度大的问题,提出一种基于自适应布谷鸟(adaptive cuckoo search,ACS)算法的Tsallis熵阈值图像分割方法,能够改善学习过程和收敛速度,减少分割时间.该方法使用Tsallis熵作为ACS的适应度函数值,实现无参数搜索过程,在搜索空间中使用当前位置的知识来自适应步长,最后使用ACS最大化Tsallis熵来获得最优阈值,得到分割图像.实验结果表明,该文方法能够有效实现图像分割,且分割时间低于粒子群优化算法、布谷鸟搜索算法和改进布谷鸟搜索算法,结构相似性(Structural Similarity, SSIM)和收敛成功率高于其他算法.
Aiming at the problem of large computational complexity in existing image segmentation algorithms, a Tsallis entropy threshold image segmentation method based on adaptive cuckoo search(ACS) algorithm has been proposed, which could improve the learning process and convergence speed, and reduce the segmentation time. In the method, Tsallis entropy has been used as the fitness function value of ACS to realize the parameterless search process. The knowledge of the current position has been used to step the length adaptively in the search space. Finally, the optimal threshold has been obtained by means of ACS to maximize the Tsallis entropy to obtain the segmentation image. The experimental results show that the proposed method can effectively achieve image segmentation, and the segmentation time is lower than particle swarm optimization algorithm, cuckoo search algorithm and improved cuckoo search algorithm, while SSIM and the convergence success rate is higher than other algorithms.
作者
黄毅英
黄河清
HUANG Yi-ying;HUANG He-qing(Department of Information Engineering,Guangxi Economic&Trade Polytechnic,Nanning 530021,China;Teachers College of Vocational and Technical Education,Guangxi Normal University,Guilin Guangxi 541004,China)
出处
《西南师范大学学报（自然科学版）》
CAS
北大核心
2020年第5期127-133,共7页
Journal of Southwest China Normal University(Natural Science Edition)
基金
广西职业教育教学改革研究项目(GXGZIG2017A036).
