Predicting optical transmission through complex scattering media from reflection patterns with deep neural networks
Kyriakos Skarsoulis, Eirini Kakkava, Demetri Psaltis
Abstract
Deep neural networks (DNNs) are used to reconstruct transmission speckle intensity patterns from the respective reflection speckle intensity patterns generated by illuminated parafilm layers. The dependence of the reconstruction accuracy on the thickness of the sample is examined for different illumination patterns of various feature sizes. High reconstruction accuracy is obtained even for large parafilm thicknesses, for which the memory effect of the sample is vanishingly small. The generalization capability of the DNN is also studied for unseen scatterers of the same type.
Topics & Concepts
Speckle patternOpticsReflection (computer programming)Transmission (telecommunications)Sample (material)Feature (linguistics)Artificial neural networkScatteringGeneralizationComputer scienceMaterials scienceIntensity (physics)Artificial intelligencePattern recognition (psychology)PhysicsTelecommunicationsMathematicsMathematical analysisPhilosophyLinguisticsThermodynamicsProgramming languageRandom lasers and scattering mediaOptical Coherence Tomography ApplicationsPhotoacoustic and Ultrasonic Imaging