2011_基于拉普拉斯谱的医学图像配准算法_夏云.pdfVIP

第37 卷 第14 期 计 算 机 工 程 2011 年7 月 Vol.37 No.14 Computer Engineering July 2011 ·图形图像处理· 文章编号：1000—3428(2011)14—0231—02 文献标识码：A 中图分类号：TP391.41 基于拉普拉斯谱的医学图像配准算法 a,b a,b a,b a,b a,b 夏 云 ，梁 栋 ，鲍文霞 ，徐 慧 ，颜 普 (安徽大学 a. 计算智能与信号处理教育部重点实验室；b. 电子信息工程学院，合肥 230039) 摘 要：提出一种基于拉普拉斯谱的医学图像配准算法，通过将谱图理论应用到医学图像配准中并引入特征向量，以达到提高配准精确度 和计算效率的目的。该算法根据医学图像的解剖特征来构造拉普拉斯矩阵，通过分析拉普拉斯矩阵的谱得到匹配关系；采用射影变换模型， 计算射影矩阵；通过坐标变换和图像插值方法实现图像配准。实验结果表明，该算法与经典的最大互信息配准算法相比，提高了单传感器 和多传感器医学图像配准的精度，并且降低运算复杂度。 关键词：配准；拉普拉斯谱；图；互信息；传感器 Medical Image Registration Algorithm Based on Laplacian Spectrum XIA Yuna,b, LIANG Donga,b, BAO Wen-xiaa,b, XU Huia,b, YAN Pua,b (a. Key Laboratory of Intelligent Computing and Signal Processing, Ministry of Education; b. School of Electronic Information Engineering, Anhui University, Hefei 230039, China) 【Abstract 】The medical image registration algorithm based on Laplacian spectrum algorithm is proposed. The idea of using graph theory as medical image point registration is proposed. By eigenvector, the higher registration precision and the strong ability have been achieved. The Laplacian spectrum of the anatomy feature point sets of the two images is constructed to match. The strategy uses the homogeneous matrix, coordinate transfer and the image interpolation to register. Experimental results show that it has more efficiency and higher registration