| 基于改进PSO算法的Otsu快速多阈值图像分割 |
| |
| 引用本文： | 彭正涛,方康玲,苏志祁.基于改进PSO算法的Otsu快速多阈值图像分割[J].现代电子技术,2011,34(6):10-14. |
| |
| 作者姓名： | 彭正涛 方康玲 苏志祁 |
| |
| 作者单位： | 武汉科技大学,信息科学与工程学院,过程控制与信息技术研究室,湖北,武汉,430081 |
| |
| 基金项目： | 湖北省教育委员会基金项目 |
| |
| 摘 要： | 为了确定图像分割的最佳阈值，基于改进粒子群优化算法，提出了一种快速多阈值图像分割方法。首先引入独立峰值将直方图划分为若干区域，然后在各个区域使用最大类间方差法得到优化的目标函数，用具有非均匀变异特性和雁群飞行启示的线性递减惯性权重粒子群算法对目标函数进行优化，得到分割的最佳阈值，并用该阈值对图像进行分割。将分割结果与常规最大类问方差法的多阈值分割结果相比较，证明该算法不仅可以正确地实现图像分割，并可使分割速度大大提高。
|
| 关 键 词： | 图像分割 粒子群算法 非均匀变异 线性递减惯性权重 独立峰值 多闽值 最大类间方差 |
Fast Multilevel Threshold Method for Image Segmentation Based on Improved Particle Swarm Optimization and Maximal Variance |
| |
| PENG Zheng-tao,FANG Kang-ling,SU Zhi-qi.Fast Multilevel Threshold Method for Image Segmentation Based on Improved Particle Swarm Optimization and Maximal Variance[J].Modern Electronic Technique,2011,34(6):10-14. |
| |
| Authors: | PENG Zheng-tao FANG Kang-ling SU Zhi-qi |
| |
| Affiliation: | PENG Zheng-tao,FANG Kang-ling,SU Zhi-qi (Department of Process Control and Information Technology,College of Information Science and Engineering,Wuhan University of Science and Technology,Wuhan 430081,China) |
| |
| Abstract: | To determine the optimal threshold in image segmentation,a new multilevel threshold method based on improved particle swarm optimization(PSO) is proposed.Firstly,the histogram was divided into several areas by the conception of independent peaks.Secondly,the maximum between-class variance(MV) method was used to get the optimization object function in each area.Thirdly,the object function was optimized by the non-uniform mutation and geese-LDW PSO,the optimal thresholds was obtained,and the image was segment... |
| |
| Keywords: | image segmentation PSO non-uniform mutation linear descend inertia weight independent peaks multilevel threshold maximal between-class variance |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
