• Corpus ID: 15887871

The End is Nigh: Generic Solving of Text-based CAPTCHAs

@inproceedings{Bursztein2014TheEI,
  title={The End is Nigh: Generic Solving of Text-based CAPTCHAs},
  author={Elie Bursztein and Jonathan Aigrain and Angelique Moscicki and John C. Mitchell},
  booktitle={Workshop on Offensive Technologies},
  year={2014}
}
The effectiveness and universality of the results suggests that combining segmentation and recognition is the next evolution of catpcha solving, and that it supersedes the sequential approach used in earlier works.
View Paper
cdn.elie.net
135 Citations
Highly Influential Citations
20
Background Citations
64
Methods Citations
45
Results Citations
6

135 Citations

Breaking Text-based CAPTCHAs using Average Vertical Partition

This paper proposes a simple but effective attack on text-based CAPTCHA that uses machine learning to solve the segmentation and recognition problems simultaneously and casts serious doubt on the security of existing text- based CAPTCHAs.

Designing a Text-based CAPTCHA Breaker and Solver by using Deep Learning Techniques

This work consists of a dynamic approach that is proposed to predict Text-based CAPTCHAs that challenges the supposition that they cannot be solved by computers.

Contour Based Deep Learning Engine to Solve CAPTCHA

A deep neural network architecture is presented to extract text from CAPTCHA images on various platforms using convolutional neural network based architecture design instead of the traditional methods ofCAPTCHA detection using image processing segmentation modules.

A generative vision model that trains with high data efficiency and breaks text-based CAPTCHAs

This work introduces recursive cortical network (RCN), a probabilistic generative model for vision in which message-passing–based inference handles recognition, segmentation, and reasoning in a unified manner and outperforms deep neural networks on a variety of benchmarks while being orders of magnitude more data-efficient.

A Generic Solver Combining Unsupervised Learning and Representation Learning for Breaking Text-Based Captchas

A generic solver combining unsupervised learning and representation learning to automatically remove the noisy background of captchas and solve text-based captchAs is proposed and outperforms state-of-the-art by delivering a higher accuracy on various captcha schemes.

I am Robot: (Deep) Learning to Break Semantic Image CAPTCHAs

A comprehensive study of reCaptcha is conducted, and a novel low-cost attack that leverages deep learning technologies for the semantic annotation of images is designed, which is extremely effective and applies to the Facebook image captcha.

SURVEY PAPER ON DESIGNING IMAGE BASED CAPTCHA USING MACHINE LEARNING

A novel image-based Captcha known as Style Area Captcha (SACaptcha) is proposed, which depends on semantic data understanding, pixel-level segmentation and deep learning techniques for increasing the security purpose.

Low-Cost Breaking of a Unique Chinese Language CAPTCHA Using Curriculum Learning and Clustering

A convolutional neural network was created to evaluate the likelihood of a region in the image belonging to an inverted character and is used with a feature map and clustering to identify potential locations of inverted characters in a distorted image.

I'm Not a Human: Breaking the Google reCAPTCHA

This paper conducts a comprehensive study of reCaptcha, and designs a novel low-cost attack that leverages deep learning technologies for the semantic annotation of images and is extremely effective.

An End-to-End Attack on Text CAPTCHAs

Experimental results prove that the anti-segmentation principle can be completely broken under deep learning attacks without any segmentation or preprocessing steps in contrast to commonly held beliefs.
...

54 References

A low-cost attack on a Microsoft captcha

This paper presents new character segmentation techniques of general value to attack a number of text CAPTCHAs, including the schemes designed and deployed by Microsoft, Yahoo and Google.

Breaking reCAPTCHA: A Holistic Approach via Shape Recognition

This work analyzes three recent generations of reCAPTCHA and presents an algorithm that is capable of solving at least 5% of the challenges generated by these versions, and proposes a machine learning algorithm that virtually eliminates the distortion.

Breaking reCAPTCHAs with Unpredictable Collapse: Heuristic Character Segmentation and Recognition

A novel approach for automatic segmentation and recognition of reCAPTCHA in Web sites based on CAPTCHA image preprocessing with character alignment, morphological segmentation with three-color bar character encoding and heuristic recognition is presented.

What's up CAPTCHA?: a CAPTCHA based on image orientation

A new CAPTCHA which is based on identifying an image's upright orientation is presented, which is language-independent, does not require text-entry, and employs another domain forCAPTCHA generation beyond character obfuscation.

Breaking Audio CAPTCHAs

This work analyzed the security of current audio CAPTCHAs from popular Web sites by using AdaBoost, SVM, and k-NN, and achieved correct solutions for test samples with accuracy up to 71%.

Recognizing objects in adversarial clutter: breaking a visual CAPTCHA

Efficient methods based on shape context matching are developed that can identify the word in an EZ-Gimpy image with a success rate of 92%, and the requisite 3 words in a Gimpy image 33% of the time.

Asirra: a CAPTCHA that exploits interest-aligned manual image categorization

A CAPTCHA that asks users to identify cats out of a set of 12 photographs of both cats and dogs, and two novel algorithms for amplifying the skill gap between humans and computers that can be used on many existing CAPTCHAs are described.

The Robustness of Google CAPTCHAs

We report a novel attack on two CAPTCHAs that have been widely deployed on the Internet, one being Google's home design and the other acquired by Google (i.e. reCAPTCHA). With a minor change, our…

Enhanced CAPTCHAs: Using Animation to Tell Humans and Computers Apart

Animated CAPTCHAs are sealed against the Laundry attack by adding a dimension not used so far: animation, which ensures that unsuspected visitors will provide answers that will be useless on the attacker's side.

The Failure of Noise-Based Non-continuous Audio Captchas

Decaptcha's performance on actual observed and synthetic CAPT CHAs indicates that such speech CAPTCHAs are inherently weak and, because of the importance of audio for various classes of users, alternative audio CAPTChAs must be developed.
...