摘要: |
本文基于遗传算法在图像相关匹配中的具体应用,提出了包括新的初始种群的选择策略以及调节收敛与发散因子、防止近亲繁殖的改进遗传策略,通过对实际图像的相关匹配仿真,证明了这些措施能有效提高遗传算法寻优可信度与寻优精度。 |
关键词: 图像处理 图像相关匹配 遗传算法 非遍历寻优 仿真 |
DOI:10.3969/j.issn.1001-893X. |
修订日期:2002-08-09 |
基金项目: |
|
A Kind of Fast Image Correlation Matching Technology Based on Improved Genetic Algorithms |
|
() |
Abstract: |
Based on the abroad application of genetic algorithms to the image correlation matching, a completely new strategy is proposed for the selection of initial population. Also, a genetic strategy is put forward, namely, modifying both the convergence factor and the divergence factor and preventing from inbreeding. The experimental results show that these strategies are effective in improving the precision and confidence of genetic algorithms. |
Key words: Image processing,Image correlation matching,Genetic algorithm,Non-ergodic optimization,Simulation |