SCRD-Net: A Deep Convolutional Neural Network Model for Glaucoma Detection in Retina Tomography

Figure 2

Structures of three kinds of deep networks. (a) Singlemodality network: data samples of a single feature are processed; (b) merge network: data samples of two features are processed together; (c) split-merge network: data samples of two features are processed separately in the bottom layers and jointly in the middle layers.

(a)

(b)

(c)