Image-based CAPTCHAs Research Articles

Adversarial Captcha

CAPTCHAs, tests to separate humans from machines, serve as a method to verify whether an interaction is being performed by a human or a computer, were developed in reaction to the vulnerability of computer networks to intrusions by programmers using bots and computer attack programs. The most popular Captcha scheme is the Text Captcha because of how simple it is to create and use. However, the presumed security of Captchas has been undermined by hackers and programmers, making websites open to attack.Given that attack speeds are generally moderate, usually ranging from two to five seconds per image and that this is not a serious worry, text captchas are nevertheless extensively utilized. This study proposes a novel image-based Captcha called Style Area Captcha (SACaptcha), which is built on deep learning techniques, pixel-level segmentation, and semantic data understanding.The proposed SACaptcha emphasises using deep learning to create neural networks neural networks image-based Captchas. Acquiring methods to enhance security and captcha systems are all covered in this paper.It compares the proving that text-based Captchas are insecure. Key words: deep, text-based, security, and captcha Convolutional image-based learning

Extended Research on the Security of Visual Reasoning CAPTCHA

CAPTCHA is an effective mechanism for protecting computers from malicious bots. With the development of deep learning techniques, current mainstream text-based and traditional image-based CAPTCHAs have been proven to be insecure. Therefore, a major effort has been directed toward developing new CAPTCHAs by utilizing some other hard Artificial Intelligence (AI) problems. Recently, some commercial companies (Tencent, NetEase, Geetest, etc.) have begun deploying a new type of CAPTCHA based on visual reasoning to defend against bots. As a newly proposed CAPTCHA, it is therefore natural to ask a fundamental question: are visual reasoning CAPTCHAs as secure as their designers expect? This paper explores the security of visual reasoning CAPTCHAs. We proposed a modular attack and evaluated it on six different real-world visual reasoning CAPTCHAs, which achieved overall success rates ranging from 79.2% to 98.6%. The results show that visual reasoning CAPTCHAs are not as secure as anticipated; this latest effort to use novel, hard AI problems for CAPTCHAs has not yet succeeded. Then, we summarize some guidelines for designing better visual-based CAPTCHAs, and based on the lessons we learned from our attacks, we propose a new CAPTCHA based on commonsense knowledge (CsCAPTCHA) and show its security and usability experimentally.

Designing Image Based Captcha using Machine Learning

Captchas, or Completely Automated Public Turing Tests to Tell Computers and Humans Apart, were created in response to programmers' ability to breach computer networks via computer attack programmes and bots. Because of its ease of development and use, the Text Captcha is the most well-known Captcha scheme. Hackers and programmers, on the other hand, have weakened the assumed security of Captchas, leaving websites vulnerable to assault. Text Captchas are still widely used since it is assumed that the attack speeds are moderate, typically two to five seconds for each image, and that this is not considered a significant concern. Style Area Captcha (SACaptcha) is a revolutionary image-based Captcha suggested in this paper, which relies on semantic data comprehension, pixel-level segmentation, and deep learning approaches. The suggested SACaptcha highlights the creation of image-based Captchas utilising deep learning techniques for boosting the security purpose, demonstrating that text Captchas are no longer secure.

Empirical studies to investigate the usability of text- and image-based CAPTCHAs

This article reports the findings of two empirical studies investigating the usability of image-based and text-based CAPTCHAs that authenticate users accessing the web. The first investigated four image-based CAPTCHAs, Asirra, ESP-PIX, SQ-PIX and IMAGINATION, with twenty participants interacting with five of each type on a desktop computer. The participants then completed the computer system usability questionnaire (CSUQ), the NASA Task Load Index (NASA-TLX), and a final post-test subjective questionnaire ranking the CAPTCHAs based on preference. The results found statistically significant differences among the CAPTCHAs for all dependent variables except task completion time, with participants preferring Asirra and ESP-PIX. The second compared the usability of these two with the two most commonly used text-based CAPTCHAs, ReCAPTCHA and Google's CAPTCHA. After receiving training on Asirra and ESP-PIX, twenty-four participants completed 10 trials with each type on a desktop computer, subsequently completing the CSUQ and the NASA-TLX followed by a retrospective think-aloud session. The results of this second study indicated that Google's CAPTCHA and ReCAPTCHA took the least amount of time to complete. The workload associated with Asirra and ReCAPTCHA was lower than that for ESP-PIX and Google's CAPTCHA. Perceived overall usability tended to be highest for Asirra and ReCAPTCHA. Participant comments indicated that Google's CAPTCHA was too visually distorted and ESP-PIX had commonalities between the images that weren't immediately obvious, factors perhaps contributing to their lower perceived usability scores. Overall, ReCAPTCHA is recommended for use on a desktop computer because of its high perceived usability, low completion time, and low workload.

Research on Deep Learning Techniques in Breaking Text-Based Captchas and Designing Image-Based Captcha

The ability of hackers to infiltrate computer systems using computer attack programs and bots led to the development of Captchas or Completely Automated Public Turing Tests to Tell Computers and Humans Apart. The text Captcha is the most popular Captcha scheme given its ease of construction and user friendliness. However, the next generation of hackers and programmers has decreased the expected security of these mechanisms, leaving websites open to attack. Text Captchas are still widely used, because it is believed that the attack speeds are slow, typically two to five seconds per image, and this is not seen as a critical threat. In this paper, we introduce a simple, generic, and fast attack on text Captchas that effectively challenges that supposition. With deep learning techniques, our attack demonstrates a high success rate in breaking the Roman-character-based text Captchas deployed by the top 50 most popular international websites and three Chinese Captchas that use a larger character set. These targeted schemes cover almost all existing resistance mechanisms, demonstrating that our attack techniques are also applicable to other existing Captchas. Does this work then spell the beginning of the end for text-based Captcha? We believe so. A novel image-based Captcha named Style Area Captcha (SACaptcha) is proposed in this paper, which is based on semantic information understanding, pixel-level segmentation, and deep learning techniques. Having demonstrated that text Captchas are no longer secure, we hope that our proposal shows promise in the development of image-based Captchas using deep learning techniques.

Evaluating the usability of CAPTCHAs on a mobile device with voice and touch input

The usability of text-based CAPTCHAs, featuring distorted letters, and image-based CAPTCHAs, featuring pictures, was explored on an Apple iPad. Five conditions were explored: Confident CAPTCHA with either voice or touch input, ESP-PIX with voice or touch input, and Google’s CAPTCHA with touch input. Usability was analyzed in terms of performance, perceived usability, workload, and preference rankings. Results showed that CAPTCHAs involving touch input scored better in almost every measure than CAPTCHAs involving voice input. In particular, Confident Touch is recommended based on preference and perceived performance, whereas ESP-PIX Touch is recommended for its short completion time. When image-based CAPTCHAs are not feasible, Google’s CAPTCHA is a satisfactory alternative based on usability ratings.

An Investigation of the Usability of Image-based CAPTCHAs

Objective: This research investigates the usability of image-based CAPTCHAs using performance, eye-gaze and subjective measures. Background: Several forms of image-based CAPTCHAs have recently been developed. Few studies have been conducted to examine their usability. Method: Twenty participants received training using four different CAPTCHAs: Asirra, ESP-PIX, SQ-PIX and IMAGINATION. The participants were then asked to complete the CAPTCHA challenge using five CAPTCHAs of each type. The participants were subsequently asked to complete the computer system usability questionnaire (CSUQ) and the NASA Task Load Index test. Upon completing all the tasks, the participants completed a final post-test subjective questionnaire ranking the CAPTCHAs based on preference. Results: Statistically significant differences among the CAPTCHAs were found for all dependent variables other than task completion time. Conclusion: Participants preferred Asirra and ESP-PIX to the other two CAPTCHAs.