针对SAR图像配准中匹配效率低、误匹配对多和配准精度差的问题,提出一种基于局部不变特征的SAR图像配准新算法.首先,使用加速分割检测特征(features from accelerated segment test,FAST)检测算法,检测SAR图像的FAST角点;使用DAISY描述...针对SAR图像配准中匹配效率低、误匹配对多和配准精度差的问题,提出一种基于局部不变特征的SAR图像配准新算法.首先,使用加速分割检测特征(features from accelerated segment test,FAST)检测算法,检测SAR图像的FAST角点;使用DAISY描述子对FAST特征进行描述,得到SAR图像不变特征。其次,采用基于KD树的欧氏距离匹配策略,实现特征点对的粗匹配;采用RANSAC算法去除误匹配,实现特征点对精匹配.然后,采用仿射变换模型,实现图像插值和图像变换,实现SAR图像粗配准。最后,建立配准精度评估反馈机制,实现配准优化.通过使用不同时相、不同工作模式HJ-1C星载SAR和不同极化、不同波段机载AIRSAR图像配准实验,提出算法与经典不变特征配准算法相比,具有适配性好、配准效率高的优点.展开更多
Currently the voxel based registration methods have been used widely such as the well known mutual information (MI). Although the accuracy of their results is plausible, the registration procedure is slow. This paper ...Currently the voxel based registration methods have been used widely such as the well known mutual information (MI). Although the accuracy of their results is plausible, the registration procedure is slow. This paper proposed some methods to rigid registration based on mutual information, aiming for an acceleration of the registration process without significantly loss of precision in the final results. The efficiency of these methods is examined by registration of CT MR and PET MR. Experimental results show that the speedup is effective and efficient. By using the fast methods, the registration of 3 D medical image could also be implemented on PC rapidly.展开更多
针对颅部三维医学图像配准计算量大、配准效率低等问题,提出了一种基于几何特征空间约束的快速配准方法。提取三维轮廓点云,提出了一种基于点云集最优拟合环的特征构造方法,并以每个特征环和每个层的质心用作特征量,通过使用迭代最近点(...针对颅部三维医学图像配准计算量大、配准效率低等问题,提出了一种基于几何特征空间约束的快速配准方法。提取三维轮廓点云,提出了一种基于点云集最优拟合环的特征构造方法,并以每个特征环和每个层的质心用作特征量,通过使用迭代最近点(Iterative Closest Point, ICP)方法完成快速配准。实验结果表明,与传统的ICP算法相比,该方法计算量小,配准精度高,配准速度快。它是一种有效的实时三维配准方法。展开更多
文摘针对SAR图像配准中匹配效率低、误匹配对多和配准精度差的问题,提出一种基于局部不变特征的SAR图像配准新算法.首先,使用加速分割检测特征(features from accelerated segment test,FAST)检测算法,检测SAR图像的FAST角点;使用DAISY描述子对FAST特征进行描述,得到SAR图像不变特征。其次,采用基于KD树的欧氏距离匹配策略,实现特征点对的粗匹配;采用RANSAC算法去除误匹配,实现特征点对精匹配.然后,采用仿射变换模型,实现图像插值和图像变换,实现SAR图像粗配准。最后,建立配准精度评估反馈机制,实现配准优化.通过使用不同时相、不同工作模式HJ-1C星载SAR和不同极化、不同波段机载AIRSAR图像配准实验,提出算法与经典不变特征配准算法相比,具有适配性好、配准效率高的优点.
文摘Currently the voxel based registration methods have been used widely such as the well known mutual information (MI). Although the accuracy of their results is plausible, the registration procedure is slow. This paper proposed some methods to rigid registration based on mutual information, aiming for an acceleration of the registration process without significantly loss of precision in the final results. The efficiency of these methods is examined by registration of CT MR and PET MR. Experimental results show that the speedup is effective and efficient. By using the fast methods, the registration of 3 D medical image could also be implemented on PC rapidly.
基金National Natural Science Foundation (61673226)Major Natural Science Research Project of Jiangsu Higher Education Institutions (18KJA470003)。
文摘针对颅部三维医学图像配准计算量大、配准效率低等问题,提出了一种基于几何特征空间约束的快速配准方法。提取三维轮廓点云,提出了一种基于点云集最优拟合环的特征构造方法,并以每个特征环和每个层的质心用作特征量,通过使用迭代最近点(Iterative Closest Point, ICP)方法完成快速配准。实验结果表明,与传统的ICP算法相比,该方法计算量小,配准精度高,配准速度快。它是一种有效的实时三维配准方法。
