Deep learning for classification of noisy QR codes

TL;DR

This paper investigates the effectiveness of deep learning models in classifying noisy QR codes, comparing their performance to traditional decoding methods to understand their limitations and potential in abstract image recognition.

Contribution

It demonstrates that deep learning models can be effective for classifying abstract images like QR codes, especially under noisy conditions, highlighting their potential beyond traditional decoding.

Findings

Deep learning models perform well on noisy QR codes.

Classical decoding methods outperform deep learning in some noise conditions.

Deep learning offers insights into abstract image classification limitations.

Abstract

We wish to define the limits of a classical classification model based on deep learning when applied to abstract images, which do not represent visually identifiable objects.QR codes (Quick Response codes) fall into this category of abstract images: one bit corresponding to one encoded character, QR codes were not designed to be decoded manually. To understand the limitations of a deep learning-based model for abstract image classification, we train an image classification model on QR codes generated from information obtained when reading a health pass. We compare a classification model with a classical (deterministic) decoding method in the presence of noise. This study allows us to conclude that a model based on deep learning can be relevant for the understanding of abstract images.