Abstract: This paper proposes an algorithm for medical image registration fused with gradient information. It extracts the complementary scale space feature points from images and calculates the gradient information around the feature points, sets the Rényi mutual information as the object function, and uses the generalized nearest-neighbor graph to estimate the Rényi entropy. Experimental results show that the proposed algorithm has better performance of speed, precision and robustness.

Key words: medical image registration, feature point, generalized nearest-neighbor graph, Rényi entropy, gradient, SIFT descriptor

摘要： 提出一种融合梯度信息的医学图像配准算法。提取图像中互补的尺度空间特征点，并计算特征点周围的梯度信息，以Rényi互信息作为目标函数，利用广义近邻图估计Rényi熵。实验结果表明，在受外界因素影响较大的情况下，该算法速度较快、准确率较高，具有较强的鲁棒性。

关键词: 医学图像配准, 特征点, 广义近邻图, Rényi熵, 梯度, SIFT描述子

CLC Number: 

• TP391.4