SIGNAL (programming language)

Quantum mechanics

Detector

Linguistics

Phase (matter)

Pixel

Philosophy

Ghost imaging

Signal-to-noise ratio (imaging)

Phase retrieval

Computer science

Photon

Physics

Pattern recognition (psychology)

Optics

Image (mathematics)

Object (grammar)

Feature (linguistics)

Artificial intelligence

Programming language

Deep learning

Artificial neural network

Noise (video)

Fourier transform

Computer vision

Physical Review Letters

Imaging systems' performance at low light intensity is affected by shot noise, which becomes increasingly strong as the power of the light source decreases. In this paper we experimentally demonstrate the use of deep neural networks to recover objects illuminated with weak light and demonstrate better performance than with the classical Gerchberg-Saxton phase retrieval algorithm for equivalent signal over noise ratio. Prior knowledge about the object is implicitly contained in the training data set and feature detection is possible for a signal over noise ratio close to one. We apply this principle to a phase retrieval problem and show successful recovery of the object's most salient features with as little as one photon per detector pixel on average in the illumination beam. We also show that the phase reconstruction is significantly improved by training the neural network with an initial estimate of the object, as opposed as training it with the raw intensity measurement.

Low Photon Count Phase Retrieval Using Deep Learning