电子学报 ›› 2018, Vol. 46 ›› Issue (10): 2384-2390.DOI: 10.3969/j.issn.0372-2112.2018.10.011

所属专题: 机器学习—特征选择

• 学术论文 • 上一篇    下一篇

一种非特征的3D图像快速刚性配准方法

周光兵, 宋华军, 吴玉兴, 任鹏   

  1. 中国石油大学(华东)信息与控制工程学院, 山东青岛 266580
  • 收稿日期:2017-03-27 修回日期:2017-09-26 出版日期:2018-10-25
    • 通讯作者:
    • 宋华军
    • 作者简介:
    • 周光兵,男.1992年生,山东泰安人.硕士研究生,现主要从事机器视觉方面的研究.E-mail:zgbupc@163.com
    • 基金资助:
    • 国家自然科学基金 (No.61305012); 国家自然科学基金 (No.61671481); 中央高校基本科研业务费专项资金 (No.18CX02109A)

A Non-Feature Fast 3D Rigid-Body Image Registration Method

ZHOU Guang-bing, SONG Hua-jun, WU Yu-xing, REN Peng   

  1. College of Information and Control Engineering, China University of Petroleum (East China), Qingdao, Shandong 266580, China
  • Received:2017-03-27 Revised:2017-09-26 Online:2018-10-25 Published:2018-10-25
    • Corresponding author:
    • SONG Hua-jun
    • Supported by:
    • National Natural Science Foundation of China (No.61305012); National Natural Science Foundation of China (No.61671481); Fundamental Research Funds for the Central Universities (No.18CX02109A)

摘要: 3D图像刚性配准旨在将一个图像映射到另一个具有相同场景的图像上,已经在医学诊断和其它领域中得到了广泛的应用.已有的方法大都基于特征点和针对特定的约束条件,带来了特征选择耗时多,随机性强,而且约束条件使用不灵活等问题.针对这些问题,提出直接使用图像灰度值的无特征3D刚性配准方法.该方法使用泰勒展开式和最小二乘法直接计算待配准图像的变换参数,并且使用较少的数据点完成快速的配准.实验结果表明,提出的算法获得较高的精度,并且使用少量的数据仍可以有效计算,这一特性使得它在大数据3D图像应用中更有吸引力.

关键词: 3D图像配准, 图像变换, 泰勒展开式

Abstract: 3D image registration (IR) aims to map one image to another image of a same scene, widely used in medical diagnosis and other applications. The existing methods mostly use feature to registration and have specific constraint condition which have many problems such as time-consuming, strong random in feature extraction and not flexible under constraint condition. For those problems, an intensity-based method for non-feature 3D rigid IR is proposed in this paper. The method uses Taylor expansion and the least squares (LS) to directly get the transformation parameters and has advantage of high processing speed with less processed data. It is shown by numerous experiments that the proposed IR method has high accuracy and only uses very small proportion data to process.

Key words: 3D image registration, image transformation, Taylor expansion

中图分类号: