Figure | All-optical machine learning decryptor (MLD) for integration on CMOS (IMAGE)

Light Publishing Center, Changchun Institute of Optics, Fine Mechanics And Physics, CAS

Figure | All-optical machine learning decryptor (MLD) for integration on CMOS

Caption

a) Through computer machine learning training, the optical MLD acquires the capability of identifying decryption keys and decoding a multitude of messages using a single decryptor element. b) The decryption system can be considered as a diffractive neural network for all-optical inference. c) Machine learning decryptors nanoprinted on CMOS chip via the GD-TPL method. (top) Photograph of the CMOS sensor with an array of 3 × 2 MLDs and 10 Euro cent coin. (middle) Scanning electron microscope image of the CMOS sensor surface with the MLD array. (bottom) Zoom-in scanning electron microscope image of a single MLD built on the CMOS sensor.

Credit

by Elena Goi, Xi Chen, Qiming Zhang, Benjamin P. Cumming, Steffen Schoenhardt?Haitao Luan and Min Gu

Usage Restrictions

None

License

Licensed content

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.