Attacking Optical Character Recognition (OCR) Systems with Adversarial Watermarks

Abstract

Optical character recognition (OCR) is widely applied in real applications serving as a key preprocessing tool. The adoption of deep neural network (DNN) in OCR results in the vulnerability against adversarial examples which are crafted to mislead the output of the threat model. Different from vanilla colorful images, images of printed text have clear backgrounds usually. However, adversarial examples generated by most of the existing adversarial attacks are unnatural and pollute the background severely. To address this issue, we propose a watermark attack method to produce natural distortion that is in the disguise of watermarks and evade human eyes' detection. Experimental results show that watermark attacks can yield a set of natural adversarial examples attached with watermarks and attain similar attack performance to the state-of-the-art methods in different attack scenarios.

9 pages, this http url http://aics.site/AICS2020/AICS20_paper_18.pdf

References in corpus (8)

• Distilling the Knowledge in a Neural Network
• Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks
• Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples
• Detecting Adversarial Samples from Artifacts
• On the (Statistical) Detection of Adversarial Examples
• LaVAN: Localized and Visible Adversarial Noise
• Attacking the Madry Defense Model with $L_1$-based Adversarial Examples
• Harmonic Adversarial Attack Method

Cited by in corpus (3)

• Generating Image Adversarial Examples by Embedding Digital Watermarks
• Universal Defensive Underpainting Patch: Making Your Text Invisible to Optical Character Recognition
• Simple Transparent Adversarial Examples