Bootstrap Resampling for Image Registration Uncertainty Estimation Without Ground Truth
Bootstrap Resampling for Image Registration Uncertainty Estimation Without Ground Truth
复制标题
DOI：
10.1109/tip.2009.2030955
复制
发表时间：
2010-01-01
影响因子：
10.6
通讯作者：
Kybic, Jan
中科院分区：
文献类型：
--
作者：
Kybic, Jan
对于没有可用地面实况数据的情况，我们解决了估计基于像素的图像配准算法的不确定性的问题，仅给出要配准的两个图像。我们的新颖方法使用引导重采样。它非常通用，适用于几乎所有基于最小化基于像素的相似性标准的配准方法；我们使用 SSD、SAD、相关性和互信息标准来演示它。我们通过实验证明，与最先进的 Cramer-Rao 绑定方法相比，引导方法可以更好地估计配准精度。此外，我们还评估了一种基于二次灵敏度分析思想的快速配准精度估计（FRAE）方法，其计算开销可以忽略不计。 FRAE 通常比 Cramer-Rao 绑定方法效果更好，但优于 bootstrap 方法。
We address the problem of estimating the uncertainty of pixel based image registration algorithms, given just the two images to be registered, for cases when no ground truth data is available. Our novel method uses bootstrap resampling. It is very general, applicable to almost any registration method based on minimizing a pixel-based similarity criterion; we demonstrate it using the SSD, SAD, correlation, and mutual information criteria. We show experimentally that the bootstrap method provides better estimates of the registration accuracy than the state-of-the-art Cramer-Rao bound method. Additionally, we evaluate also a fast registration accuracy estimation (FRAE) method which is based on quadratic sensitivity analysis ideas and has a negligible computational overhead. FRAE mostly works better than the Cramer-Rao bound method but is outperformed by the bootstrap method.