Fig. 2
From: Deep learning-based analysis of macaque corneal sub-basal nerve fibers in confocal microscopy images
The entire processing pipeline used in this study. The input image (a) is pre-processed to compensate for variation in the background illumination (b). The segmentation, performed using deep learning, generates a probability image that assigns a score between 0 and 1 to each pixel (c) depicting a pixel-wise nerve classification. The final post-processing step is to binarize and skeletonize that result (d)