基于联合直方图区域计数的时间序列图象自动配准方法

Automatic Serial Image Registration by Regional Counting in Joint-Histogram

时间序列图象的配准是医学临床科研、诊断和治疗中图象应用的必要步骤 .为了快速、准确、简便地进行医学序列图象配准 ,提出了一种新的用于时间序列图象自动配准的方法 ,该方法利用了图象的联合直方图 ,首先通过对图象做简单的阈值分割 ,将联合直方图划分为 4个区域 ,然后根据不同的配准图象数据选择定义在不同区域上的计数值作为参数计算的准则函数 .该方法设计简单、巧妙 ,以计数方法代替其他方法中大量的浮点运算 .由于准则函数具有良好的光滑特性 ,且选择 Powell算法做最优化搜索 ,因此保证了优化结果的准确性 .和其他算法相比 ,该方法大大简化了准则函数的计算 ,从而显著提高了配准优化搜索的速度 .根据实验结果及基于互信息量方法做的对比 ,证明本文提出的方法准确、简便、快速。

Registration of serial images plays an important role in clinical research, diagnosis therapy and evaluation in medicine. A novel method used for serial image registration is proposed in this paper, which is based on the joint histogram of the two images to be registered. Usually, it is very easy, simply with a properly selected threshold, to segment human tissue from the background in medical images. With the thresholds of the two image sets to be registered, the joint histogram is divided into four separate regions. The configuration of the joint histogram bears a greatly different appearance between the before registered and after registered images. This difference appearance reflects the degree of the registration. The criterion function is defined as counting in a specific region or several regions of the joint histogram, which simplifies the computation complexity greatly and speeds up the alignment process significantly. With the fine property of the criterion function, powell's direction set method in multi dimensions is chosen to carry out the optimizing process to calculate the registration parameters. The comparison of the results from both mutual information and our method shows that the new method based on segmentation and counting is a fast, simple, efficient and accurate registration method.