Deep learning for classification of noisy QR codes
This work addresses the problem of classifying noisy abstract images like QR codes, which could benefit applications in automated data retrieval, but it appears incremental as it applies existing deep learning methods to a new data type.
The researchers investigated the limitations of deep learning models for classifying abstract images by training an image classification model on QR codes generated from health pass data and comparing it with deterministic decoding methods under noisy conditions. They concluded that deep learning can be relevant for understanding abstract images, though no concrete performance numbers were provided.
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.