David R. Cheriton School of Computer Science, University of Waterloo, Waterloo, ON N2L 3G1, Canada.
IEEE Trans Image Process. 2010 May;19(5):1236-47. doi: 10.1109/TIP.2009.2039371. Epub 2009 Dec 28.
Many registration scenarios involve aligning more than just two images. These image sets-called ensembles-are conventionally registered by choosing one image as a template, and every other image is registered to it. This pairwise approach is problematic because results depend on which image is chosen as the template. The issue is particularly acute for multisensor ensembles because different sensors create images with different features. Also, pairwise methods use only a fraction of the available data at a time. In this paper, we propose a maximum-likelihood clustering method that registers all the images in a multisensor ensemble simultaneously. Experiments involving rigid-body and affine transformations show that the clustering method is more robust and accurate than competing pairwise registration methods. Moreover, the clustering results can be used to form a rudimentary segmentation of the image ensemble.
许多注册场景涉及到不仅仅是两个图像的对齐。这些图像集——称为集合——通常通过选择一个图像作为模板来注册，而其他每个图像都注册到它。这种成对的方法存在问题，因为结果取决于选择哪个图像作为模板。对于多传感器集合来说，这个问题尤其严重，因为不同的传感器会创建具有不同特征的图像。此外，成对的方法一次只使用可用数据的一小部分。在本文中，我们提出了一种最大似然聚类方法，它可以同时注册多传感器集合中的所有图像。涉及刚体和仿射变换的实验表明，聚类方法比竞争的成对注册方法更稳健和准确。此外，聚类结果可用于形成图像集合的基本分割。
