| 基于模拟退火与遗传算法结合的神经网络图像分割 |
| |
| 引用本文： | 何汉青,倪祥刚,李志俊.基于模拟退火与遗传算法结合的神经网络图像分割[J].武汉理工大学学报,2011(5). |
| |
| 作者姓名： | 何汉青 倪祥刚 李志俊 |
| |
| 作者单位： | 武汉理工大学网络信息中心;武汉理工大学自动化学院; |
| |
| 基金项目： | 中央高校基本科研业务费专项资金(2010-ZX-010) |
| |
| 摘 要： | 为解决运用Hopfield神经网络优化算法处理图像分割存在的收敛速度与局部最优的矛盾,采用模拟退火策略与遗传算法结合的优化方法来改进传统的优化算法,对迭代收敛后的Hopfield网络在局部范围内运用模拟退火遗传算法,以搜索阈值平面全局最优解,进行图像分割。实验证明,采用此方法可以得到较好的分割效果。
|
| 关 键 词： | Hopfield网络 模拟退火 遗传算法 优化 |
Simulated Annealing and Genetic Algorithms Based on Image Segment with Partially Evolved Hopfield Neural Network |
| |
| HE Han-qing,NI Xiang-gang,LI Zhi-jun.Simulated Annealing and Genetic Algorithms Based on Image Segment with Partially Evolved Hopfield Neural Network[J].Journal of Wuhan University of Technology,2011(5). |
| |
| Authors: | HE Han-qing NI Xiang-gang LI Zhi-jun |
| |
| Affiliation: | HE Han-qing1,NI Xiang-gang2,LI Zhi-jun2(1.Network Information Center,Wuhan University of Technology,Wuhan 430070,China,2.School of Automation,China) |
| |
| Abstract: | A combined optimization of genetic algorithms with simulated annealing has been applied to image segmentation with good results in this paper.The defect of Hopfield neural network is being captured by local optimal solutions,while the defect of genetic algorithms is the low speed of convergence.Both disadvantages mentioned above have been overcome here.Solutions obtained with the converged Hopfield neural network are applied to the genetic algorithm to search for the optimization on the plane of threshold v... |
| |
| Keywords: | Hopfield neural network simulated annealing genetic algorithms optimization |
| 本文献已被 CNKI 等数据库收录！ |
