Matched Longitudinal Image Pairs for Development

Validation dataset for the paper:Chen, M., Cooper, R.F., Gee, J.C., Brainard, D.H. and Morgan, J.I., 2019. Automatic longitudinal montaging of adaptive optics retinal images using constellation matching. Biomedical Optics Express, 10(12), pp.6476-6496.Matched pairs of longitudinal AO image triplets (confocal, split detection, dark field) at 30 retinal locations across 5 subjects. This dataset was used for tuning during the development stage of the algorithm. The intensity for the split detection in the earlier dataset for each subject was inverted to maintain consistent orientation with current protocols.本数据集为下述论文的验证数据集：Chen, M.、Cooper, R.F.、Gee, J.C.、Brainard, D.H. 及 Morgan, J.I. 于2019年发表于《生物医学光学快报（Biomedical Optics Express）》的论文《基于星座匹配的自适应光学视网膜图像纵向拼接（Automatic longitudinal montaging of adaptive optics retinal images using constellation matching）》。该数据集包含5名受试者视网膜30个成像位置处的纵向自适应光学（Adaptive Optics, AO）图像三联组（共焦（confocal）、分光检测（split detection）、暗场（dark field））的匹配对。本数据集用于该算法开发阶段的参数调优环节。为与当前实验方案保持一致的成像取向，已将每名受试者早期数据集中的分光检测通道图像的灰度强度进行了反转处理。