Computer Engineering and Applications ›› 2017, Vol. 53 ›› Issue (23): 171-176.DOI: 10.3778/j.issn.1002-8331.1704-0161

Previous Articles     Next Articles

Research on image edge detection based on improved ant colony algorithm

WANG Kai, ZHANG Guicang   

  1. College of Mathematics and Statistics, Northwest Normal University, Lanzhou 730070, China
  • Online:2017-12-01 Published:2017-12-14

基于改进蚁群算法的图像边缘检测研究

汪  凯,张贵仓   

  1. 西北师范大学 数学与统计学院,兰州 730070

Abstract: There are some questions that the edge is unsmooth, the noise is affected greatly, and it is easy to converge to local, when the traditional ant colony algorithm is used in edge detection. In order to improve the quality of edge detection, firstly, the algorithm in this paper determines the initial position and the heuristic matrix by combining the method of the gray gradient and the region gray mean; secondly, the weight factor is introduced to define the new probability transfer function, and the pheromone matrix is updated by the chaos algorithm and adaptive parameters. The experimental results show that the improved ant colony algorithm can effectively reduce the influence of noise on edge detection, and detect image edge that is more complete and clear.

Key words: ant colony algorithm, edge detection, weight, gradient, region gray mean, adaptive

摘要: 传统的蚁群算法应用于图像边缘检测时,会出现边缘不够平滑、受噪声影响大、易收敛于局部等问题。为了提高边缘检测的效果,将灰度梯度与区域灰度均值方法相结合,确定蚂蚁的初始位置和启发矩阵;引入权重因子定义新的概率转移函数,并通过混沌算法和自适应参数进行信息素矩阵的更新,避免过早陷入局部最优。实验结果表明,改进的蚁群算法可以有效减少噪声对边缘检测的影响,并获得更加完整和清晰的图像边缘,取得较好的效果。

关键词: 蚁群算法, 边缘检测, 权重, 梯度, 区域灰度均值, 自适应