| 基于改进遗传算法的车牌图像分割 |
| |
| 引用本文： | 李正峰,侯杰. 基于改进遗传算法的车牌图像分割[J]. 计算机工程与设计, 2005, 26(9): 2455-2457 |
| |
| 作者姓名： | 李正峰 侯杰 |
| |
| 作者单位： | 泰州职业技术学院,江苏,泰州,225300;烟台师范学院,山东,烟台,264025 |
| |
| 摘 要： | 为了解决简单遗传算法存在的早熟、收敛到全局最优解难问题,提出了适应度标定公式,提高选择压力;定义相似度概念,保留相似性差的个体,剔除相似性个体.在不增加群体规模的前提下,增加了群体的多样性.改进遗传算法能够避免简单遗传算法所存在的问题.为了解决车牌图像识别率低问题,提出基于改进遗传算法的车牌图像分割方法,进行车牌图像识别.实验证明利用此方法进行车牌图像识别效果较好.
|
| 关 键 词： | 模式识别 图像分割 车牌:遗传算法 |
| 文章编号： | 1000-7024（2005）09-2455-03 |
| 收稿时间： | 2004-08-31 |
| 修稿时间： | 2004-08-31 |
License plate image segmentation based on modified genetic algorithms |
| |
| LI Zheng-feng,HOU Jie. License plate image segmentation based on modified genetic algorithms[J]. Computer Engineering and Design, 2005, 26(9): 2455-2457 |
| |
| Authors: | LI Zheng-feng HOU Jie |
| |
| Affiliation: | 1. Taizhou Polytechnic Institute, Taizhou 225300, China; 2. Yantai Normal University, Yantai 264025, China |
| |
| Abstract: | In order to avoid prematurity and convergence out of optimized point for simple genetic algorithms, a fitness normalization formula is introduced and the press for selection of individuals is increased. The similarity is defined to increase the varieties of individuals without increasing the size of population, which can solve the problem of local optimized solution. In order to solve the problem on lower rate to recognition to license plate image, an image segmentation method based on modified genetic algorithms is also proposed, which can segment recognize fonts out of license plate perfectly. |
| |
| Keywords: | pattern recognition image segmentation license plate genetic algorithms |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
