Journal: Eye
Article Title: Ultra-wide field and new wide field composite retinal image registration with AI-enabled pipeline and 3D distortion correction algorithm
doi: 10.1038/s41433-023-02868-3
Figure Lengend Snippet: Sample Heidelberg and Optos checkerboard overlay image of RANSAC-SC ( A ), OR CNN ( B ), and OR CNN + DC ( C ) with its corresponding magnified image of alignment (a, b, and c, respectively). Images using the AI algorithms OR CNN and OR CNN + DC showed more aligned images than RANSAC-SC, as seen from the encircled vessels. RANSAC-SC Random Sample Consensus, Sample and Consensus sets, OR CNN Outlier Rejection Convolutional Neural Network, OR CNN + DC Outlier Rejection Convolutional Neural Network + Distortion Correction.
Article Snippet: Sample Heidelberg and Optos checkerboard overlay image of RANSAC-SC ( A ), OR CNN ( B ), and OR CNN + DC ( C ) with its corresponding magnified image of alignment (a, b, and c, respectively).
Techniques: