3D Reconstruction of Retinal Blood Vessels from Two Views

A 3D reconstruction of retinal blood vessel trees using two views of fundus images is presented. The problem addresses: 1) The recovery of camera-eye model parameters by an self-calibration method. The camera parameters are found via the solution of simplified Kruppa equations, based on correspondences captured by hand from four different views. 2) The extraction of blood vessels and skeletons from two fundus images. 3) The match of correspondence points of the two skeleton trees. The trees are previously labelled during the analysis of 2D binary images. 4) The lineal triangulation of matched correspondence points, and the surface modeling using diameter measurements extracted from the 2D binary images. Examples from 3D skeletons and tree surfaces reconstructed models are shown.