CAPTCHA

About: CAPTCHA is a research topic. Over the lifetime, 1304 publications have been published within this topic receiving 18987 citations. The topic is also known as: Completely Automated Public Turing test to tell Computers and Humans Apart & captcha.

Captcha identification method and apparatus, and computer device and storage medium

Abstract: Disclosed are a CAPTCHA identification method and apparatus based on data analysis, and a computer device and a storage medium. The method comprises: acquiring a target CAPTCHA picture (S1); inputting the target CAPTCHA picture into a CAPTCHA identification model based on a convolutional neural network to carry out computation to obtain a computation result, wherein the CAPTCHA identification model comprises a CNN module and an LSTM module, and in the CAPTCHA identification model, target feature extraction is carried out on the target CAPTCHA picture by using the CNN module, an extracted target feature is input into the LSTM module to carry out sequence prediction to obtain a prediction result, and the prediction result is then computed by means of a preset function to obtain the computation result (S2); and acquiring the computation result output by the CAPTCHA identification model, wherein the computation result is a CAPTCHA character corresponding to the target CAPTCHA picture (S3). Therefore, identification is directly completed from end to end, and the model has a smaller computation amount and higher identification accuracy.

A feature matching-based attack on text CAPTCHAs

Abstract: CAPTCHA is a key defense technology that improves the security of a website. Text CAPTCHA schemes are the first barrier against attacks on major websites because of simplicity and low cost. In this paper, we proposed a novel, simple, and easy-to-operate feature matching approach to attack text CAPTCHAs. Our method combines a localization network and a feature matching network to attack three real-world text CAPTCHA schemes, including Wikipedia, Baidu and JD. We used 2000 samples for our attack experiments, and the results showed that our proposed method achieved high success rates on all three schemes.

Robust image-based CAPTCHA generation using adversarial attack

Abstract: CAPTCHA (Completely Automated Public Truing test to tell Computers and Humans Apart) is a technology used to protect network security, distinguishing humans and machines by tests that humans can easily pass but machines fail. However, with the wide application of CAPTCHAs, many cracking methods have emerged. Specifically, Deep Learning (DL) have played an important role in in destroying CAPTCHA, which brings a great threat to the security of CAPTCHAs. Therefore, it is of great significance to improve the security of the CAPTCHAs against cracking methods based on deep learning technology. This paper focuses on image-based CAPTCHAs which is the most widely used type, and proposes an image-based CAPTCHA generation method based on adversarial examples. Starting from the actual application scenario of CAPTCHA, we improve it on the basis of I-FGSM (Iterative Fast Gradient Sign Method) by randomly transforming the input samples and performing secondary processing to adversarial noise. Experiments show that our method can effectively improve the robustness and transferability of the CAPTCHA.

CAPTCHAs in reality: What might happen if they are easily broken?

Abstract: This research paper explains the potential impact when CAPTCHAs become easy for attackers to solve. A literature review has been done on the different types of CAPTCHAs as well as the different methods present to break the CAPTCHAs. But data regarding the effect that CAPTCHAs have had when they have been attacked and broken is scarce. CAPTCHAs are deployed mainly to secure sites or apps but there are particular disadvantages which make them not only less beneficial to the owners of the sites or apps but also to the humans who need to solve them. These disadvantages are also in the form of vulnerabilities which prove beneficial to the attackers who can easily make use of the vulnerabilities to break the CAPTCHAs. It has been found that the impact of these attacks, in reality, has not been mentioned in previous studies. The impact prompts for advanced CAPTCHAs which solve the problems with the CAPTCHAs we have today.

Do Regional Variations Affect the CAPTCHA User Experience? A Comparison of CAPTCHAs in China and the United States

Abstract: Systems worldwide deploy CAPTCHAs as a security mechanism to protect from unauthorized automated access. Typically, the effectiveness of CAPTCHAs is evaluated based on their resilience against bots. User perceptions of the interactive experience and effectiveness of CAPTCHAs have received less attention, especially for comparing the variations of CAPTCHAs presented in different regions across the world. As the first step toward filling this gap, we conducted semi-structured interviews with ten participants fluent in Chinese and English to investigate whether user perceptions are affected by variations in CAPTCHAs presented in China and the United States, respectively. We found notable differences in the perceived user experience and effectiveness across the different CAPTCHA types, but not across regional variations of the same type. Our findings point to a number of avenues for making the CAPTCHA user experience more universal and inclusive.