| 引用本文: | 金闳奇,简川霞,赵荣丽.基于互信息和混合算法的印刷图像配准[J].包装工程,2018,39(13):194-198. |
| |
|
| |
|
|
| 本文已被：浏览 2910次 下载 1766次 |
码上扫一扫！ |
|
|
| 基于互信息和混合算法的印刷图像配准 |
|
金闳奇1, 简川霞1, 赵荣丽2,3
|
|
1.广东工业大学，广州 510006;2.1.广东工业大学，广州 510006;3.2.广东省创新方法与决策管理系统重点实验室，广州 510006
|
|
| 摘要: |
| 目的 为了提高印刷图像配准的精度，提出一种基于混合搜索算法的图像配准方法。方法 首先求取图像的归一化互信息，然后利用GA算法（遗传算法）进行全局搜索，得出粗配准参数；最后，利用Powell算法进行局部寻优，得出精配准参数。结果 混合算法的配准结果与只用单一Powell搜索算法或只用单一GA搜索算法相比，在各个几何变换方向上得到了更小的配准误差。结论 与GA算法和Powell算法相比，文中建议的混合算法配准精确度更高、速度更快。 |
| 关键词: 归一化互信息 遗传算法 Powell算法 图像配准 |
| DOI：10.19554/j.cnki.1001-3563.2018.13.032 |
| 分类号:TS801.3 |
| 基金项目:国家自然科学基金（51605101）；广东省数控一代机械产品创新应用示范工程专项资金项目（2013B011301023） |
|
| Printing Images Registration Based on Mutual Information and Hybrid Algorithm |
|
JIN Hong-qi1, JIAN Chuan-xia1, ZHAO Rong-li2,3
|
|
1.Guangdong University of Technology, Guangzhou 510006, China;2.1.Guangdong University of Technology, Guangzhou 510006, China;3.2.Guangdong Provincial Key Laboratory of Innovation Method and Decision Management System, Guangzhou 510006, China
|
| Abstract: |
| The work aims to propose an image registration method based on hybrid search algorithm, so as to improve the accuracy of printing image registration (IR). Firstly, the normalized mutual information of the image was calculated, and then the global search was done by GA (genetic algorithm) to obtain the rough registration parameters. Finally, the Powell algorithm was used for local optimization to obtain the precise registration parameters. Compared with the single Powell or GA search algorithm, the registration results of hybrid algorithm had smaller registration error in each geometric transformation direction. The proposed hybrid algorithm registration is much more accurate and fabulously faster than GA algorithm and Powell algorithm. |
| Key words: normalized mutual information genetic algorithm Powell algorithm image registration |
|
