Gong Chen, Brunton Steven L, Schowengerdt Brian T, Seibel Eric J
University of Washington, Department of Mechanical Engineering, Seattle, Washington, United States.
Magic Leap, Plantation, Florida, United States.
J Med Imaging (Bellingham). 2021 Sep;8(5):054002. doi: 10.1117/1.JMI.8.5.054002. Epub 2021 Sep 29.
Handling low-quality and few-feature medical images is a challenging task in automatic panorama mosaicking. Current mosaicking methods for disordered input images are based on feature point matching, whereas in this case intensity-based registration achieves better performance than feature-point registration methods. We propose a mosaicking method that enables the use of mutual information (MI) registration for mosaicking randomly ordered input images with insufficient features. Dimensionality reduction is used to map disordered input images into a low dimensional space. Based on the low dimensional representation, the image global correspondence can be recognized efficiently. For adjacent image pairs, we optimize the MI metric for registration. The panorama is then created after image blending. We demonstrate our method on relatively lower-cost handheld devices that acquire images from the retina , kidney , and bladder phantom, all of which contain sparse features. Our method is compared with three baselines: AutoStitch, "dimension reduction + SIFT," and "MI-Only." Our method compared to the first two feature-point based methods exhibits 1.25 ( microscope dataset) to two times ( retina dataset) rate of mosaic completion, and MI-Only has the lowest complete rate among three datasets. When comparing the subsequent complete mosaics, our target registration errors can be 2.2 and 3.8 times reduced when using the microscopy and bladder phantom datasets. Using dimensional reduction increases the success rate of detecting adjacent images, which makes MI-based registration feasible and narrows the search range of MI optimization. To the best of our knowledge, this is the first mosaicking method that allows automatic stitching of disordered images with intensity-based alignment, which provides more robust and accurate results when there are insufficient features for classic mosaicking methods.
在自动全景拼接中，处理低质量和特征较少的医学图像是一项具有挑战性的任务。当前针对无序输入图像的拼接方法基于特征点匹配，而在这种情况下，基于强度的配准比特征点配准方法具有更好的性能。我们提出了一种拼接方法，该方法能够使用互信息（MI）配准来拼接特征不足的随机排序输入图像。降维用于将无序输入图像映射到低维空间。基于低维表示，可以有效地识别图像全局对应关系。对于相邻图像对，我们优化MI度量以进行配准。然后在图像融合后创建全景图。我们在相对低成本的手持设备上展示了我们的方法，这些设备采集来自视网膜、肾脏和膀胱模型的图像，所有这些图像都包含稀疏特征。我们的方法与三个基线进行了比较：AutoStitch、“降维+尺度不变特征变换（SIFT）”和“仅MI”。与前两种基于特征点的方法相比我们的方法在拼接完成率方面表现出1.25倍（显微镜数据集）到两倍（视网膜数据集）的提升，并且“仅MI”在三个数据集中的完成率最低。在比较后续完整的拼接图时，使用显微镜和膀胱模型数据集时，我们的目标配准误差可以降低2.2倍和3.8倍。使用降维提高了检测相邻图像的成功率，这使得基于MI的配准可行，并缩小了MI优化的搜索范围。据我们所知，这是第一种允许使用基于强度对齐的无序图像自动拼接的方法，当经典拼接方法的特征不足时，该方法能提供更稳健和准确的结果。
