Robust and efficient single-pixel image classification with nonlinear optics

Abstract

We present a hybrid image classifier by feature-sensitive image upconversion, single pixel photodetection, and deep learning, aiming at fast processing of high-resolution images. It uses partial Fourier transform to extract the images’ signature features in both the original and Fourier domains, thereby significantly increasing the classification accuracy and robustness. Tested on the Modified National Institute of Standards and Technology handwritten digit images and verified by simulation, it boosts accuracy from 81.25% (by Fourier-domain processing) to 99.23%, and achieves 83% accuracy for highly contaminated images whose signal-to-noise ratio is only ${-}{{17}}\;{\rm{dB}}$. Our approach could prove useful for fast lidar data processing, high-resolution image recognition, occluded target identification, and atmosphere monitoring.

© 2021 Optical Society of America

Single-pixel pattern recognition with coherent nonlinear optics

Ting Bu, Santosh Kumar, He Zhang, Irwin Huang, and Yu-Ping Huang
Opt. Lett. 45(24) 6771-6774 (2020)

Single-pixel ptychography

Meng Li, Liheng Bian, Guoan Zheng, Andrew Maiden, Yang Liu, Yiming Li, Jinli Suo, Qionghai Dai, and Jun Zhang
Opt. Lett. 46(7) 1624-1627 (2021)

Robust 17 W single-pass second-harmonic-generation at 532 nm and relative-intensity-noise investigation

Clément Dixneuf, Germain Guiraud, Hanyu Ye, Yves-Vincent Bardin, Mathieu Goeppner, Giorgio Santarelli, and Nicholas Traynor
Opt. Lett. 46(2) 408-411 (2021)

Optical incoherent synthetic aperture imaging by superposition of phase-shifted optical transfer functions

Mani Ratnam Rai and Joseph Rosen
Opt. Lett. 46(7) 1712-1715 (2021)

Compact long-range single-photon imager with dynamic imaging capability

Peng-Yu Jiang, Zheng-Ping Li, and Feihu Xu
Opt. Lett. 46(5) 1181-1184 (2021)