[18p-P06-1~11] FS Focused Session "AI Electronics" (Poster)

[18p-P06-9] Simulation of image processing for Magneto-Optical Diffraction Deep Neural Networks

〇Reo Akagawa¹, Hotaka Sakaguchi¹, Hirofumi Nonaka², Hiroyuki Awano³, Fatima Zahara Chafi¹, Takayuki Ishibashi¹ (1.Nagaoka Univ. Tech., 2.Aichi Inst. Tech., 3.Toyota Tech. Inst.)

Keywords:

Deep Neural Network,optics,Magneto optical effect

In recent years, Deep Neural Networks have garnered significant attention and have been applied in various fields such as image processing and natural language processing. However, the increasing processing speed and power consumption pose significant challenges. To address these issues, we propose a magneto-optical effect-based diffractive deep neural network (MO-D²NN) as a physical device solution. In this study, leveraging the capability of MO-D²NN to rapidly and parallel process two-dimensional image data, we investigated the performance of an image processing task that classifies handwritten numbers images as either even or odd.