Figure 1.

A schematic visualization of the deep neural network for single-particle-imaging inversion. The neural network is implemented using an architecture composed entirely of convolutional, maximum pooling and upsampling layers. In the network, Conv. refers to convolution, LRLU refers to the leaky rectified linear unit and BN refers to batch normalization. There are two channels in the final output layer, for the reconstruction of the amplitude and the phase of the particle. All activations are related to the LRLU, except for the final convolutional layer which uses RLU activations.