Fooling OCR Systems with Adversarial Text Images
Machine Learning
2018-02-16 v1 Artificial Intelligence
Cryptography and Security
Computer Vision and Pattern Recognition
Abstract
We demonstrate that state-of-the-art optical character recognition (OCR) based on deep learning is vulnerable to adversarial images. Minor modifications to images of printed text, which do not change the meaning of the text to a human reader, cause the OCR system to "recognize" a different text where certain words chosen by the adversary are replaced by their semantic opposites. This completely changes the meaning of the output produced by the OCR system and by the NLP applications that use OCR for preprocessing their inputs.
Cite
@article{arxiv.1802.05385,
title = {Fooling OCR Systems with Adversarial Text Images},
author = {Congzheng Song and Vitaly Shmatikov},
journal= {arXiv preprint arXiv:1802.05385},
year = {2018}
}