Robust and Efficient Single-Pixel Image Classificationwith Nonlinear Optics

Abstract

We present a hybrid image classifier by mode-selective image upconversion, single pixel photodetection, and deep learning, aiming at fast processing a large number of pixels. It utilizes partial Fourier transform to extract the signature features of images in both the original and Fourier domains, thereby significantly increasing the classification accuracy and robustness. Tested on the MNIST handwritten digit images, it boosts the accuracy from 81.25% to 99.23%, and achieves an 83% accuracy for highly contaminated images whose signal-to-noise ratio is only -17 dB. Our approach could prove useful for fast lidar data processing, high resolution image recognition, occluded target identification, atmosphere monitoring, and so on.

5 pages, 5 figures

References in corpus (17)

• Deep Learning in Neural Networks: An Overview
• Parallel convolution processing using an integrated photonic tensor core
• All Optical Neural Network with Nonlinear Activation Functions
• Large-Scale Optical Neural Networks based on Photoelectric Multiplication
• Low Photon Count Phase Retrieval Using Deep Learning
• Theory of neuromorphic computing by waves: machine learning by rogue waves, dispersive shocks, and solitons
• A Photonic In-Memory Computing primitive for Spiking Neural Networks using Phase-Change Materials
• Optical machine learning with incoherent light and a single-pixel detector
• Polaritonic neuromorphic computing outperforms linear classifiers
• Image-free real-time classification of fast moving objects using 'learned' spatial light modulation and a single-pixel detector
• Large-scale Ising Emulation with Four-Body Interaction and All-to-All Connection
• Deep learning reconstruction of ultrashort pulses from 2D spatial intensity patterns recorded by an all-in-line system in a single-shot
• Classification and Verification of Online Handwritten Signatures with Time Causal Information Theory Quantifiers
• Mode-selective Image Upconversion
• Single-Pixel Pattern Recognition with Coherent Nonlinear Optics
• Mode Selective Image Upconversion over Turbulence
• Fast Partial Fourier Transform

Cited by in corpus (4)

• Gigahertz-rate random speckle projection for high-speed single-pixel image classification
• Observation of Distinct Phase Transitions in a Nonlinear Optical Ising Machine
• Optical convolutional neural network with atomic nonlinearity
• Dense Associative Memory in a Nonlinear Optical Hopfield Neural Network