广义Jensen-Schur测度在医学图像配准中的应用  

Application of generalized Jensen-Schur measure in medical image registration

作　　者：胡顺波[1] 王广泰[1] 刘常春[2] 邵鹏[3] 

机构地区：[1]临沂师范学院物理系,山东临沂276005 [2]山东大学控制科学与工程学院,济南250061 [3]阿尔伯塔大学放射与诊断影像学系,加拿大阿尔伯塔T6G2E1

出　　处：《计算机应用》2009年第5期1441-1445,共5页journal of Computer Applications

基　　金：国家863计划项目(2006AA02Z4D9)

摘　　要：使用互信息或归一化互信息测度进行医学图像配准时,由于噪声、模态、插值等影响,测度函数存在许多局部极值,收敛范围较窄,有可能导致误配准。为了克服上述缺点,定义了广义Jensen-Schur测度,利用巴特沃思函数对自变量值的非线性压缩功能,成功地消除了PV插值伪极值点。从测度曲线光滑性、抗噪鲁棒性、收敛性能方面,对四种新构造的广义Jensen-Schur测度、互信息和归一化互信息进行了比较和分析。实验结果表明,新构造的JS22和JS23测度在以上三个方面的性能都优于其他测度。For the influences of noise, interpolation and image modality, the medical image registration method based on mutual information or normalized mutual information would cause local extrema, small convergence area, and even inaccurate registration. A new generalized Jensen-Sehur measure was defined, which used ＂nonlinear increasing＂ of butterworth function to eliminate false extrema. Four new generalized Jensen-Schur measures, mutual information and normalized mutual information were analyzed and compared by applying them to rigid registration. The results of tests show that the new constructed JS22 and JS23 measures outperform other measures in noise immunity and convergence, and eliminating false extrema caused by PV interpolation.

关 键 词：图像配准 广义Jensen—Schur测度 互信息 PV插值 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]