Image classification using collective optical modes of an array of nanolasers

TL;DR

This paper demonstrates that an 8x8 nanolaser array can perform binary image classification with 98% accuracy by exploiting collective optical modes and simple training strategies, advancing optical computing capabilities.

Contribution

It introduces a novel approach using collective optical modes of nanolaser arrays for image classification, achieving high accuracy with simple training.

Findings

Achieved 98% success rate in binary image recognition.

Utilized symmetry properties of collective modes for classification.

Demonstrated potential for high-density, low-power optical computing.

Abstract

Recent advancements in nanolaser design and manufacturing open up unprecedented perspectives in terms of high integration densities and ultra-low power consumption, making these devices ideal for high-performance optical computing systems. In this work we exploit the symmetry properties of the collective modes of a nanolaser array for binary image classification. The implementation is based on a 8x8 array, and relies on the activation of a collective optical mode of the array, the so-called "zero mode", under spatially modulated pump patterns. We demonstrate that a simple training strategy allows us to achieve an overall success rate of 98% in binary image recognition.