Fourier-space Diffractive Deep Neural Network

作者：Tao Yan, Jiamin Wu, Tiankuang Zhou, Hao Xie, Feng Xu, Jingtao Fan, Lu Fang, Xing Lin, Qionghai Dai · 发表于：Physical Review Letters · 年份：2019 · DOI：10.1103/physrevlett.123.023901 · 被引用次数：411 · 研究领域：Neural Networks and Reservoir Computing、Photonic and Optical Devices、Optical Network Technologies

In this Letter we propose the Fourier-space diffractive deep neural network ($\mathrm{F}\text{\ensuremath{-}}\mathrm{D}^{2}\mathrm{NN}$) for all-optical image processing that performs advanced computer vision tasks at the speed of light. The $\mathrm{F}\text{\ensuremath{-}}\mathrm{D}^{2}\mathrm{NN}$ is achieved by placing the extremely compact diffractive modulation layers at the Fourier plane or both Fourier and imaging planes of an optical system, where the optical nonlinearity is introduced from ferroelectric thin films. We demonstrated that $\mathrm{F}\text{\ensuremath{-}}\mathrm{D}^{2}\mathrm{NN}$ can be trained with deep learning algorithms for all-optical saliency detection and high-accuracy object classification.