A needle-based deep-neural-network camera

TL;DR

This paper introduces a needle-sized camera using a cannula as the main optic, combined with deep neural networks to reconstruct images, generate depth maps, and classify data, enabling compact imaging with privacy benefits.

Contribution

It presents a novel needle-based camera system integrated with deep neural networks for image reconstruction, depth estimation, and classification, advancing miniaturized imaging technology.

Findings

Successful reconstruction of color and grayscale images.

Generation of accurate depth maps from trained DNNs.

Effective classification of EMNIST dataset with and without image reconstruction.

Abstract

We experimentally demonstrate a camera whose primary optic is a cannula (diameter=0.22mm and length=12.5mm) that acts a lightpipe transporting light intensity from an object plane (35cm away) to its opposite end. Deep neural networks (DNNs) are used to reconstruct color and grayscale images with field of view of 180 and angular resolution of ~0.40. When trained on images with depth information, the DNN can create depth maps. Finally, we show DNN-based classification of the EMNIST dataset without and with image reconstructions. The former could be useful for imaging with enhanced privacy.