[LCT5-1(Invited)]All–Optical Diffractive Deep Neural Networks Enabled by Liquid Crystal
Liu Liu2,1, Quanzhou Long1, Fion Sze Yan Yeung2, Man Chun Tseng2, *Wanlong Zhang1(1. Shenzhen University (China), 2. Hong Kong University of Science and Technology (Hong Kong))
Keywords:
Liquid Crystal,Optical Neural Network,Photo-induced Alignment
In this paper, we introduce a novel approach utilizing a 5–layer all–optical D2NN constructed with photo–induced liquid crystal alignment technology to create LC–based tunable phase retarders, experimentally achieving a classification accuracy of 89% with 500 random hand-written digits from the test MNIST dataset.
