| 基于SIFT算法的可见光图像与红外图像配准 |
| |
| 引用本文： | 周文理,金施群.基于SIFT算法的可见光图像与红外图像配准[J].黑龙江科技学院学报,2011,21(2):121-123,156. |
| |
| 作者姓名： | 周文理 金施群 |
| |
| 作者单位： | 合肥工业大学,仪器科学与光电工程学院,合肥,230009 |
| |
| 摘 要： | 为实现可见光图像与红外图像的有效配准,结合两类图像的特点,提出基于特征的SIFT算法,并进行道路和人脸的可见光图像与红外图像配准实验。结果表明,SIFT算法对纹理结构简单、光滑图像提取的特征点较少;随着阈值提高,可见光图像与红外图像的匹配点数目增加,匹配稳定性降低。该算法可用于可见光图像与红外图像配准。
|
| 关 键 词： | 图像配准 SIFT 红外 可见光 |
Visible/infrared image registration based on SIFT |
| |
| ZHOU Wenli,JIN Shiqun.Visible/infrared image registration based on SIFT[J].Journal of Heilongjiang Institute of Science and Technology,2011,21(2):121-123,156. |
| |
| Authors: | ZHOU Wenli JIN Shiqun |
| |
| Institution: | (School of Instrument Science & Opto-electronic Engineering,Hefei University of Technology,Hefei 230009,China) |
| |
| Abstract: | Aimed at an effective registration of visible/infrared image, this paper proposes features- based SIFT algorithm, combined with the characteristics of two types of image, and describes an experi- ment focused on registration of road and face visible/infrared image. The result indicates that SIFT algo- rithm allows fewer features to be extracted from simple texture structure and smooth image. The increasing threshold means a greater number of the image matches and a lower matching stability. This algorithm can apply to visible/infrared image registration. |
| |
| Keywords: | SIFT |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
