Image Recognition Using Oxide Semiconductor Crossbar Memristors with Implementation of Slit Detection and Local Autonomous Learning

Yuta Takishita

doi:https://doi.org/10.36463/idw.2020.0973

17:55 〜 18:10

[AIS1/PRJ1-4L] Image Recognition Using Oxide Semiconductor Crossbar Memristors with Implementation of Slit Detection and Local Autonomous Learning

*Yuta Takishita¹, Mutsumi Kimura^1,2, Yasuhiko Nakashima¹ (1. Nara Institute of Science and Technology(Japan), 2. Ryukoku University(Japan))

crossbar-memristor, neural-network, neuromorphic, local-autonomous-learning

https://doi.org/10.36463/idw.2020.0973

In this study, we simulated an image recognition system using visual cortex layers and xbar memristor as a full connection layer. The simulation results show that the best accuracy was 69% when the MNIST datasets were used, and it was 23% when the CIFAR-10 datasets were used.

Proceedings of the International Display Workshops Volume 27 (IDW '20)

[AIS1/PRJ1-4L] Image Recognition Using Oxide Semiconductor Crossbar Memristors with Implementation of Slit Detection and Local Autonomous Learning