What is it about?
Text-based CAPTCHA is the most popularly deployed method for preventing bot attacks and attracted many researchers to focus on its security. In this paper, we summarize existing text-based CAPTCHAs and their corresponding attacks, propose a unified attack framework, and perform an experimental investigation on the effects of existing attacks on text-based CAPTCHA schemes. This paper also provides an open-access CAPTCHA dataset that contains 22 different CAPTCHA sample sets.
Featured Image
Photo by Petter Lagson on Unsplash
Why is it important?
We provide a new taxonomy for text-based CAPTCHAs based on their resistance mechanisms and an open-access dataset containing 22 different CAPTCHA sample sets. We establish attacks under a consistent benchmark using the same datasets and metrics. Our findings show that text-based CAPTCHAs are vulnerable to both one-stage and multistage attacks, and most existing resistance mechanisms are ineffective.
Perspectives
I hope this article can provide people who are interested in CAPTCHA with more useful information about text CAPTCHA security and attacks. Writing this article was a great pleasure as it has co-authors with whom I have had long standing collaborations.
Ping Wang
Xidian University
Read the Original
This page is a summary of: An Experimental Investigation of Text-based CAPTCHA Attacks and Their Robustness, ACM Computing Surveys, January 2023, ACM (Association for Computing Machinery),
DOI: 10.1145/3559754.
You can read the full text:
Contributors
The following have contributed to this page
