| 基于改进SIFT算法的超声图像配准研究 |
| |
| 引用本文： | 郑芳,李颖,陆雪松,高智勇.基于改进SIFT算法的超声图像配准研究[J].现代科学仪器,2012(1):31-35. |
| |
| 作者姓名： | 郑芳 李颖 陆雪松 高智勇 |
| |
| 作者单位： | 中南民族大学生物医学工程学院,武汉,450074 |
| |
| 基金项目： | 国家自然科学基金项目(61002046) |
| |
| 摘 要： | 针对SIFT算法对医学超声图像提取关键点数目较少的缺点,本文提出了采用HARRIS角点改进SIFT算法的方法,有效提取更具有结构意义的关键点数目,提高超声图像配准精度。此外,为进一步提高配准精度,增加一步互信息算法对图像进行精配。实验结果表明,本文算法是一种快速、高精度、全自动的医学图像配准算法,并可有效地提高配准精度。
|
| 关 键 词： | SIFT HARRIS 互信息 图像配准 |
Research on Ultrasound Image Registration Method Based on Improved SIFT Algorithm |
| |
| Zheng Fang,Li Ying,Lu Xuesong,Gao Zhiyong.Research on Ultrasound Image Registration Method Based on Improved SIFT Algorithm[J].Modern Scientific Instruments,2012(1):31-35. |
| |
| Authors: | Zheng Fang Li Ying Lu Xuesong Gao Zhiyong |
| |
| Affiliation: | (Institute of Biological and Medical Engineering,South-Central University for Nationalities,Wuhan 430074) |
| |
| Abstract: | To overcome the shortcoming of the Scale Invariant Feature Transform(SIFT) algorithm in ultrasound image registration,this paper proposes an improved method.The method integrated HARRIS algorithm into SIFT so that it can extract structural and significant pionts in original image.Moreover,further registration was implemented by mutual information method to achieve more accurate result.Experiment results showed that the proposed algorithm is more quickly,accuracte and full-automatic. |
| |
| Keywords: | SIFT HARRIS Mutual Information Image Registration |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
