Anuj Thakur

Bio: Anuj Thakur is an academic researcher from Birla Institute of Technology and Science. The author has contributed to research in topics: CAPTCHA. The author has an hindex of 1, co-authored 1 publications receiving 4 citations.

A Reading Oriented Overlapping Text based CAPTCHA

Abstract: CAPTCHA is a security technique used widely over the Internet to verify whether the end user is a human. Text-based CAPTCHAs are the most commonly used. The security of text-based CAPTCHA depends on the ability to resist the segmentation and recognition attacks. The Reading Oriented Overlapping Text (ROOT) based CAPTCHA, introduced in this article, consists of overlapping alphanumeric letters both hand written and computer generated along with a reading pattern scheme that helps to use the same CAPTCHA multiple number of times. The proposed ROOT based CAPTCHA generation algorithm is discussed along with its resistivity to known attacks.

A machine learning attack against variable-length Chinese character CAPTCHAs

Abstract: CAPTCHA (Completely Automated Public Turing test to tell Computer and Human Apart) is widely used as a standard security mechanism to protect resources on websites. Among various kinds of CAPTCHAs, the text-based CAPTCHA is the most popular scheme, which consists of English letters, Arabic digits and other character sets, such as Chinese characters. Due to the large quantity of Chinese characters and complicated character structure, it is difficult for bots to crack Chinese character CAPTCHAs. Thus, Chinese character CAPTCHAs have been widely applied in China. Nevertheless, effective offensive approaches are necessary to help CAPTCHA designers find security vulnerabilities to improve defense mechanisms. To deal with variable-length Chinese character CAPTCHAs with noises, an automatic attacking approach is proposed, which includes preprocessing, character segmentation and character recognition. For character recognition, two methods are proposed: MGLCR (Multi-scale Gabor and Logistic regression based CAPTCHA Recognition) and CCR (Convolutional neural network based CAPTCHA Recognition). MGLCR extracts features by multi-scale Gabor filters and classifies characters with logistic regression. CCR extracts features and recognize characters automatically with CNN (Convolutional Neural Network). Experimental results show that the proposed approaches are efficient in attacking variable-length Chinese character CAPTCHAs with noises. The pros and cons of proposed MGLCR and CCR methods are discussed, which outperform state-of-the-art methods. Besides, the proposed methods could achieve satisfactory results in breaking the mixed character CAPTCHAs which consist of English letters, Arabic digits, Chinese characters and mathematical operators.

CAPTCHA: A Security Review

Abstract: Completely Automated Public Turing test to tell Computers and Human Apart(CAPTCHA) is an automated test that humans can pass, but current computer programs could not pass. The main purpose of CAPTCHA is to block automated scripts that are posted spam content everywhere they can. In this paper, various types of CAPTCHA techniques has been discussed. There is a constant require to enhance current CAPTCHAs and to develop new CAPTCHAs in order to provide security against newly generated programs which can create thousands of e-mail accounts used for stuff online polls with ballots, malicious purposes, and develop worms and viruses contained in emails.

CAPTCHA Techniques of Secure Web Authentication: A Survey

Abstract: Today, web security is the foremost concern due to the requirement of a validation and verification process that not only authenticates the end user but also prevents the data from various malicious programs. Due to the huge use of internet and electronic devices, the computer bots try to access the authentic data resided at the servers. The CAPTCHA is generally meant for authentication of the websites during login, for securing and getting access of data. A CAPTCHA challenge is kept for doing the conformation whether the process is initiated by an authentic user or by an attacker. If the challenge is successfully chased, it is signified as a human or else a computer bot, and the access will be provided to the right user. Our aim is to make a natural interface between authentic user and web servers so that the communication between the entities will take place in a secured manner. This paper presents a brief summary on CAPTCHA and its necessity in current scenarios. We have discussed a study on various types of CAPTCHAs available, along with their methodologies and the implementation details which help to secure the web data. The paper, moreover, presents some of the important advantages, disadvantages and applications of these techniques and technologies used in different research in addition to the different future scope of CAPTCHAs.

CAPTCHA: A Systematic Review

Abstract: With the massive growth of accessing Internet-based services by users, there is a need to limit services use to human beings only rather than intelligent bots. The biggest challenge in front of researchers is to determine incoming requests are from benign users or bots. A CAPTCHA (Completely Automated Public Turing Test to tell Computers and Humans Apart) is a basic tool for preventing access to Internet-based services by intelligent bots. It plays a vital role in numerous security applications for denying automatic registration conducted by bots. This paper presents a systematic review of different types of CAPTCHAs and primarily focuses on the benefits of "text-based CAPTCHAs" over other schemes. Further, this paper highlights the major deficiencies of the text-based CAPTCHA schemes.