Building Segmentation Based Human-Friendly Human Interaction Proofs (HIPs)

Chellapilla, Kumar; Larson, Kevin; Simard, Patrice; Czerwinski, Mary

doi:10.1007/11427896_1

Cited by 108 publications

(107 citation statements)

References 9 publications

Supporting

Mentioning

105

Contrasting

Unclassified

Order By: Relevance

“…By design, any system that breaks reCAPTCHA is a step towards better OCR software. 4 Our results indicate that Shape Contexts could be a valuable fallback solution in the domain of character recognition.…”

Section: Discussionmentioning

confidence: 76%

“…This is absolutely necessary for security because, as Chellapilla et al discuss in [5], recognizing single characters is a solved problem where computers even outperform humans. The task of segmentation on the other hand, where one is interested in partitioning a string in terms of a connected component into its individual characters, is still considered fairly resistant against automated attacks [4]. Shape Contexts.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Breaking reCAPTCHA: A Holistic Approach via Shape Recognition

Baecher

Büscher

Fischlin

et al. 2011

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Abstract.CAPTCHAs are small puzzles which should be easily solvable by human beings but hard to solve for computers. They build a security cornerstone of the modern Internet service landscape, deployed in essentially any kind of login service, allowing to distinguish authorized human beings from automated attacks. One of the most popular and successful systems today is reCAPTCHA. As many other systems, reCAPTCHA is based on distorted images of words, where the distortion system evolves over time and determines different generations of the system.In this work, we analyze three recent generations of reCAPTCHA and present an algorithm that is capable of solving at least 5% of the challenges generated by these versions. We achieve this by applying a specialized variant of shape contexts proposed by Belongie et al. to match entire words at once. In order to handle the ellipse shaped distortions employed in one of the generations, we propose a machine learning algorithm that virtually eliminates the distortion. Finally, an improved shape matching strategy allows us to use word dictionaries of a reasonable size (with approximately 20,000 entries).

show abstract

Section: Discussionmentioning

confidence: 76%

Section: Related Workmentioning

confidence: 99%

Breaking reCAPTCHA: A Holistic Approach via Shape Recognition

Baecher

Büscher

Fischlin

et al. 2011

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

show abstract

“…This is because any practical scheme necessitates that humans must be able to correctly solve the challenges with a high success rate, while at the same time the probability that a computer program can correctly solve it must be very small. It has been stated that the design of a robust CAPTCHA must somehow capitalize on the difference in ability between humans and current computer programs [5]. With advances in research areas like computer vision and machine learning, it is debatable whether or not it is possible to design a CAPTCHA that is easy for humans but difficult for computer programs, and as such this remains an open question.…”

Section: Design Issuesmentioning

confidence: 99%

“…Its use has been effective, to a certain extent, in deterring the automated abuse of online services intended for humans [5]. As examples, CAPTCHAs have been effective in preventing malicious bot programs from signing up for thousands of free email accounts and in deterring bots from sending out thousands of spam messages every minute [6].…”

Section: Introductionmentioning

confidence: 99%

Text-based CAPTCHAs over the years

Chow

Susilo

2017

IOP Conf. Ser.: Mater. Sci. Eng.

View full text Add to dashboard Cite

“…Specification-wise, CAPTCHA do not need to be 100% effective at identifying software bots. A design goal for a CAPTCHA could be to prevent bots from having a success rate greater than 0.01% [6]. Since CAPTCHA use increases the cost of a software robot, the CAPTCHA can still be effective, as long as this increased cost remains higher than the cost of using a human.…”

Section: Captchamentioning

confidence: 99%

Audio CAPTCHA for SIP-Based VoIP

Soupionis¹,

Tountas²,

Gritzalis³

2009

Emerging Challenges for Security, Privacy and Trust

View full text Add to dashboard Cite

Abstract. Voice over IP (VoIP) introduces new ways of communication, while utilizing existing data networks to provide inexpensive voice communications worldwide as a promising alternative to the traditional PSTN telephony. SPam over Internet Telephony (SPIT) is one potential source of future annoyance in VoIP. A common way to launch a SPIT attack is the use of an automated procedure (bot), which generates calls and produces audio advertisements. In this paper, our goal is to design appropriate CAPTCHA to fight such bots. We focus on and develop audio CAPTCHA, as the audio format is more suitable for VoIP environments and we implement it in a SIP-based VoIP environment. Furthermore, we suggest and evaluate the specific attributes that audio CAPTCHA should incorporate in order to be effective, and test it against an open source bot implementation.

show abstract

Building Segmentation Based Human-Friendly Human Interaction Proofs (HIPs)

Cited by 108 publications

References 9 publications

Breaking reCAPTCHA: A Holistic Approach via Shape Recognition

Breaking reCAPTCHA: A Holistic Approach via Shape Recognition

Text-based CAPTCHAs over the years

Audio CAPTCHA for SIP-Based VoIP

Contact Info

Product

Resources

About