A reading-based CAPTCHA, called ‘ScatterType,’ designed to resist character–segmentation attacks, is described. Its challenges are pseudorandomly synthesized images of text strings rendered in machine-print typefaces: within each image, characters are fragmented using horizontal and vertical cuts, and the fragments are scattered by vertical and horizontal displacements. This scattering is desig...

CAPTCHA stands for Completely Automated Public Turing Test to Tell Computer and Human Apart. It is the test that human can pass easily but computer cannot. For example, humans can read distorted text, but current computer programs cannot. For many years, CAPTCHAs have proven very useful for many reputable, Web-based email and application service providers, including social networking sites and ...

Over the last decade, it has become well-established that a captcha’s ability to withstand automated solving lies in the difficulty of segmenting the image into individual characters. The standard approach to solving captchas automatically has been a sequential process wherein a segmentation algorithm splits the image into segments that contain individual characters, followed by a character rec...

We present a black-box attack against an already deployed CAPTCHA that aims to protect a free service delivered using the Internet. This CAPTCHA, referred to as ‘‘Math CAPTCHA’’ or ‘‘QRBGS CAPTCHA’’, requests the user to solve a mathematical problem in order to prove human. We study significant problems both in its design and its implementation, and how those flaws can be used to completely sol...

“CAPTCHA” stands for Completely Automated Public Turing Test[1] to Tell Computers and Humans Apart[2]. Due to exponential growth of internet, security of web application has become a vital issue and many web applications facing a threat of web bots also known as internet Robot is an automated script which executes over the web forms and occupy web spaces and thus increases network traffic. The ...

This paper studies the effect of individual differences on user performance related to text-recognition CAPTCHA challenges. In particular, a text-recognition CAPTCHA mechanism was deployed in a three-month user study to investigate the effect of individuals’ different cognitive processing abilities, targeting on speed of processing, controlled attention and working memory capacity toward effici...

A feature based relation classification approach is presented in this paper. We aimed to exact relation candidates from Wikipedia texts. A probabilistic and a semantic relatedness features are employed with other linguistic information for the purpose. The experiments show that, relation classification using the proposed relatedness features with surface information like word and part-of-speech...

CAPTCHA stands for “Completely Automated Public Turing test to tell Computers and Humans Apart” and has received much attention since it first began appearing on websites. It requires the deciphering of distorted texts, mostly in English which is something that computers still cannot do well. It is also helpful in preventing the abuse of online services. The current text-based CAPTCHA requires ...

CAPTCHA is now commonly used as standard security technology to tell computers and humans apart. The most widely deployed CAPTCHAs are text-based schemes. In this paper, we document how we have broken such a text-based scheme which uses the “connecting characters together (CCT)” principle. CAPTCHAs of this type can be classified into three types: CAPTCHA with overlap but no noise arcs; CAPTCHA ...