应用显著纹理特征的医学图像配准

针对传统的基于几何度量的配准方法无法配准存在局部变形的医学器官的问题,提出了应用显著纹理特征的经典迭代最近点(ICP)医学图像配准算法。该方法借鉴主动外观模型(AAM)思想对医学图像的显著纹理特征建模,将显著性强的特征点赋予较大权重,率先配准。在传统基于空间距离的图像配准基础上加入显著纹理距离。然后,模拟格式塔心理学提出的人类视觉认知过程,使用线性递减的权重平衡两种“距离”度量方式。该算法前期主要根据几何距离取得整体配准效果,后期依赖图像纹理特征使存在局部变形位置的特征点也能精确配准。最后,在腹腔肝脏图像上进行实验。实验结果表明该算法取得了较好的配准效果,准确率达78.82%,比其他几种流行算法提高了22.22%,且对图像的旋转变化不敏感。提出的算法基本解决了存在局部变形医学器官图像的配准问题,达到了精度高、鲁棒性强的配准效果。

Abstract

Traditional registration methods based on geometric measurement can not match the medical image with local deformation. To solve the problem,an improved Iterative Closest Points (ICP) algorithm about human visual cognitive process is proposed base on the salient texture.Firstly,the method establishes the model for the salient texture feature of a medical image based on Active Appearance Model(AAM )algorithm,and it gives the feature point with more salient for a larger weight to complete the image match in the first step. Then,it introduces the salient texture distance to the traditional space distance. By simulating the human visual cognitive process proposed by Gestalt,the linear decreasing weight is used to balance the two kinds of distance measuring methods. With the algorithm,a whole registration is obtained by the geometric distance in the early stage. On the other hand,the feature points of local deformation are accurately registrated with the texture features in the later stage. At last,a series of experiments are performed on real live images.The experiment results show that the algorithm can get a good matching result,and the registration accuracy is 7882%,increasing by 22.22% as compared with those of other popular algorithms. The experimental results also show that it is not sensitive to the rotation of the images. It concludes that the algorithm solves the registration problems of local deformation in medical organs and achieves higher precision and stronger robustness.

参考文献

[1] 王雷,高欣,崔学理,等. 基于灰度距离融合的 2D/3D 刚性配准［J］. 光学精密工程,2014,22(10): 2815-2824.WANG L,GAO X,CUI X L,et al.. 2D/3D rigid registration by integrating intensity distance ［J］. Opt. Precision Eng.,2014,22(10): 2815-2824.(in Chinese)

[2] OLIVEIRA FPM,TAVARES JMRS. Medical image registration: a review ［J］. Computer Methods in Biomechanics and Biomedical Engineering,2014,17(2): 73-93.

[3] SENIN N,COLOSIMO B M,PACELLA M. Point set augmentation through fitting for enhanced ICP registration of point clouds in multisensor coordinate metrology ［J］. Robotics and Computer-Integrated Manufacturing,2013,29(1): 39-52.

[4] 陈高. 基于特征与灰度的医学图像配准方法［D］. 厦门: 厦门大学,2009.CHEN G. Medical Image Registration Method Based on Characteristic and Gray ［D］. Xiamen: Xiamen Uuiversity,2009.(in Chinese)

[5] DONG J M,PENG Y X,YING S H,et al.. LieTrICP: An improvement of trimmed iterative closest point algorithm ［J］. Neurocomputing,2014,140: 67-76.

[6] 孙军华,谢萍,刘震,等. 基于分层块状全局搜索的三维点云自动配准［J］. 光学精密工程,2013,21(1): 174-180.ZHANG J H,XIE P,LIU ZH,et al.. Automatic 3D point cloud registration based on hierarchical block global search ［J］. Opt. Precision Eng.,2013,21(1): 174-180.(in Chinese)

[7] SHARP G C,LEE S W,WEHE D K. ICP registration using invariant features ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(1): 90-102.

[8] KASE K,MAKINOUCHI A,NAKAGAWA T,et al.. Shape error evaluation method of free-form surfaces ［J］. Computer-Aided Design,1999,31(8): 495-505.

[9] SCHTZ C,JOST T,HGLI H. Multi-feature matching algorithm for free-form 3D surface registration ［C］.Fourteenth International Conference on Pattern Recognition,Piscataway,NJ. IEEE,1998(2): 982-984.

[10] JOHNSON A E,BING KANG S. Registration and integration of textured 3D data ［J］. Image and Vision Computing,1999,17(2): 135-147.

[11] 孙宁. 认知过程的格式塔感知基础［J］.考试周刊,2012 (12): 21-22.SUN N. Cognitive processes based on gestalt perception ［J］. Examination weekly,2012(12): 21-22.(in Chinese)

[12] DAMAS S,CORDN O,SANTAMARA J. Medical image registration using evolutionary computation: An experimental survey ［J］. IEEE Computational Intelligence Magazine,2011,6(4): 26-42.

[13] 金左轮,韩静,张毅,等. 基于纹理显著性的微光图像目标检测［J］. 物理学报,2014,63(6): 69501.JIN Z L,HAN J,ZHANG Y,et al.. Low light level image target detection based on texture saliency［J］. Acta Physica Sinica,2014,63(6): 69501.(in Chinese)

[14] 赵宏伟,陈霄,刘萍萍,等. 视觉显著目标的自适应分割［J］. 光学精密工程,2013,21(2): 531-538.ZHAO H W,CHEN X,LIU P P,et al.. Adaptive segmentation for visual salient object ［J］. Opt. Precision Eng.,2013,21(2): 531-538.(in Chinese)

[15] 肖德贵,辛晨,张婷,等. 显著性纹理结构特征及车载环境下的行人检测［J］. 软件学报,2014,25(3): 675-689.XIAO D G,XIN CH,ZHANG T,et al.. Saliency texture structure descriptor and its application in pedestrian detection ［J］.Journal of Software,2014,25(3): 675-689.(in Chinese)

[16] CHENG X,SRIDHARAN S,SARAGIH J,et al. Rank minimization across appearance and shape for AAM ensemble fitting ［C］.IEEE International Conference on Computer Vision (ICCV),Piscataway,2013: 577-584.

[17] 刘丽霞. 图像纹理特征研究和比较.［D］北京: 北京邮电大学,2011.LIU L X. Study and comparison on image texture featur ［D］. Beijing: Beijing University of Posts and Telecommunications,2011.(in Chinese)

[18] RIAZ F,HASSAN A,REHMAN S,et al.. Texture classification using rotation-and scale-invariant Gabor texture features ［J］. IEEE Signal Processing Letters,2013,20(6): 607-610.

[19] HAGHIGHAT M,ZONOUZ S,ABDEL-MOTTALEB M. Identification using encrypted biometrics ［C］.Computer Analysis of Images and Patterns,Springer Berlin Heidelberg,2013: 440-448.

[20] 晓娟,李忠科,王先泽,等. 基于特征点和改进ICP的三维点云数据配准算法［J］. 传感器与微系统,2012(9): 116-118,122.XIAO J,LI ZH K,WANG X Z,et al.. Research of 3D point cloud data registration algorithms based on feature points and improved ICP ［J］. Transducer and Microsystem Technologies,2012(9): 116-118,122.(in Chinese)

[21] 张红颖. 医学图像配准算法研究［D］.天津: 天津大学,2007.ZHANG H Y. Research on medical image registration algorithm ［D］. Tianjin: Tianjin University,2007.(in Chinese)

[22] COOPER G R J. The antialiased textural analysis of aeromagnetic data ［J］. Computers & Geosciences,2009,35(3): 586-591.

王红玉, 冯筠, 崔磊, 贺小伟, 邱实. 应用显著纹理特征的医学图像配准[J]. 光学精密工程, 2015, 23(9): 2656. WANG Hong-yu, FENG Jun, CUI Lei, HE Xiao-wei, QIU Shi. Medical image registration based on salient texture[J]. Optics and Precision Engineering, 2015, 23(9): 2656.

应用显著纹理特征的医学图像配准

关于本站 Cookie 的使用提示

全站搜索

应用显著纹理特征的医学图像配准

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索