Multilevel Image Thresholding Using Tsallis Entropy and Cooperative Pigeon-inspired Optimization Bionic Algorithm
Multilevel Image Thresholding Using Tsallis Entropy and Cooperative Pigeon-inspired Optimization Bionic Algorithm
复制标题
使用 Tsallis 熵和协作鸽子优化仿生算法进行多级图像阈值处理
DOI：
10.1007/s42235-019-0109-1
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Zhang Xiaofeng
中科院分区：
文献类型：
--
作者：
Wang Yun;Zhang Guangbin;Zhang Xiaofeng
多阈值分割是一种简单有效的图像分割方法。在本文中，我们提出了一种新的多级阈值的方法，使用合作鸽子启发的优化算法与动态距离阈值（CPIOD）的提高适用性和实用性的最佳阈值技术。首先，我们在鸽子启发优化算法的地图和罗盘算子中引入合作行为，以克服“维数灾难”，帮助算法快速收敛。然后，通过引入距离阈值，保持种群的多样性，增加种群的生命力，避免局部最优。Tsallis熵被用来作为目标函数来评估所考虑的灰度图像的最佳阈值。采用四幅基准图像测试了CPIOD算法和其他三种优化算法在多阈值分割问题中的性能和稳定性。四种优化算法的分割结果表明，CPIOD算法不仅能得到更高质量的分割结果，而且具有更好的稳定性。
Multilevel thresholding is a simple and effective method in numerous image segmentation applications. In this paper, we propose a new multilevel thresholding method that uses cooperative pigeon-inspired optimization algorithm with dynamic distance threshold (CPIOD) for boosting applicability and the practicality of the optimum thresholding techniques. Firstly, we employ the cooperative behavior in the map and compass operator of the pigeon-inspired optimization algorithm to overcome the “curse of dimensionality” and help the algorithm converge fast. Then, a distance threshold is added to maintain the diversity of the pigeon population and increase the vitality to avoid local optimization. Tsallis entropy is used as the objective function to evaluate the optimum thresholds for the considered gray scale images. Four benchmark images are applied to test the property and the stability of the proposed CPIOD algorithm and three other optimization algorithms in multilevel thresholding problems. Segmentation results of four optimization algorithms show that CPIOD algorithm can not only get higher quality segmentation results, but also has better stability.