Abstract
In the past decades, due to the low design cost and easy maintenance, text-based CAPTCHAs have been extensively used in constructing security mechanisms for user authentications. With the recent advances in machine/deep learning in recognizing CAPTCHA images, growing attack methods are presented to break text-based CAPTCHAs. These machine learning/deep learning-based attacks often rely on training models on massive volumes of training data. The poorly constructed CAPTCHA data also leads to low accuracy of attacks. To investigate this issue, we propose a simple, generic, and effective preprocessing approach to filter and enhance the original CAPTCHA data set so as to improve the accuracy of the previous attack methods. In particular, the proposed preprocessing approach consists of a data selector and a data augmentor. The data selector can automatically filter out a training data set with training significance. Meanwhile, the data augmentor uses four different image noises to generate different CAPTCHA images. The well-constructed CAPTCHA data set can better train deep learning models to further improve the accuracy rate. Extensive experiments demonstrate that the accuracy rates of five commonly used attack methods after combining our preprocessing approach are 2.62% to 8.31% higher than those without preprocessing approach. Moreover, we also discuss potential research directions for future work.
Highlights
Automated Public Turing tests to tell Computers and Humans Apart, abbreviated as CAPTCHA, is a kind of test that automatically distinguishes human and robot operations
We evaluate the proposed preprocessing approach by integrating it into five commonly used machine learning-based attack methods. ese commonly used attack methods include convolutional neural network (CNN), support vector machine (SVM), decision tree (DT), random forest (RF), and logistic regression (LR)
(2) We perform comprehensive ablation experiments by combining our approach in five commonly used machine/deep learning methods, including CNN, SVM, DT, RF, and LR. e experiment results confirm that our approach can significantly improve the high accuracy of the general attack methods, indicating the wide applicability of our approach
Summary
Automated Public Turing tests to tell Computers and Humans Apart, abbreviated as CAPTCHA, is a kind of test that automatically distinguishes human and robot operations. Nearly 55% of websites use text-based CAPTCHAs as security and identification mechanisms, far exceeding other types, due to the low development and maintenance costs. For this reason, we mainly focus on the study of textbased CAPTCHAs in this paper. As an evolution of machine learning, deep learning is capable of making accurate decisions in the field of image recognition [12] For this reason, recent research efforts began to explore deep learning-based attack methods to solve text-based CAPTCHAs with a high-accurate accuracy rate [7, 8]. It is not worth wasting a lot of time for us to manually label these types of data
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have