Automated public turing test

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.



A CAPTCHA(Completely Automated Public Turing Test to Tell Computers and Humans Apart) is a challenge-response test most often placed within web forms to determine whether the user is human.It is also known as HIP(Human Interaction Proof). The purpose of CAPTCHA is to block form submissions by spambots, which are automated scripts that post spam content everywhere they can.

Due to the failure of various previous versions of captcha which were easily hacked,Niloy Mitra and his colleagues from IIT,Delhi came up with idea of using animated ink blot images. The team's new system uses so-called "emerging images" - seemingly random assortments of blotches from which a coherent image emerges after a few seconds.(Barras Colin,"Animated Ink_blot Images keep unwanted blots at Bay",Nov 2009)

His team's new design uses so-called "emerging images" - seemingly random assortments of blotches from which a coherent image emerges after a few seconds .

To produce the emerging image, they have developed an algorithm that identifies key features within an original image and converts them into an array of ink blots or "splats". It then removes a number of the splats to make it harder for bots to reconstruct the original shape - while leaving enough information for a human brain to do so.

The number of splats and the noise in the background can be tweaked to make the emerging image easier or harder to spot. Tests with 310 volunteers showed that 98 per cent could recognise over 80 per cent of the emerging images at the easy setting, taking 6.4 seconds on average to do so." 03 November 2009 by Colin Barras

Moni Naor was the first person to theorize a list of ways to verify that a request comes from a human and not a bot.[4] Primitive CAPTCHAs seem to have been developed in 1997 by Andrei Broder, Martin Abadi, Krishna Bharat, and Mark Lillibridge to prevent bots from adding URLs to their search engine.[5] In order to make the images resistant to OCR (Optical Character Recognition), the team simulated situations that scanner manuals claimed resulted in bad OCR. In 2000, Luis von Ahn, Manuel Blum, Nicholas J. Hopper, and John Langford coined the term 'CAPTCHA', improved and publicized the notion, which included any program that can distinguish humans from computers. They invented multiple examples of CAPTCHAs, including the first CAPTCHAs to be widely used, which were those adopted by Yahoo!.


With the increasing number of hacking incidents relating old CAPTCHA techniques, The need to develop new innovative and fool-proof designs has been found extremely important all across the world. Many famous web-sites like Google , Yahoo! And even Microsoft have been victims of CAPTCHA failure and to ensure complete cyber-security, research on Animated Ink-Blot CAPTCHA images has been pushed to second gear. Computer Scientists from all across the world are collaborating with each other to come up with a solution to the ever-increasing menace of CAPTCHA breakdowns.

According to Larry Seltzer in [1] the main purpose of CAPTCHA or Completely Automated Public Turing Test ( see D4 ) to tell Computers and Humans Apart is to present a user with a question that only a human can decipher ( see D1 ) and answer correctly. He also mentions that with the drastic development of robots and computer malware CAPTCHAS have proved inefficient in ensuring security.

Niloy Mitra, a computer scientist at the Indian Institute Of Technology, Delhi [2] also mentions that the use of Animated Ink-Blot CAPTCHAs can make it harder for bots (see D3 ) to solve and much easier for human users to handle. He along with his colleague also devised that the ink-blot images can separate the bots from the humans. However, according to him it sometimes makes it difficult for humans to decipher as well.

Pass Rates for these CAPTCHAS could be a problem, says Luis von Ahn at Carnegie Mellon University in Pittsburgh, Pennsylvania, co-creator of the written captchas found on the web today. His ReCaptcha update to the technology was recently bought by Google.[2]

Mitra and Cohen-Or in [2] also mention that adding another elements could make the emerging image design much better. According to them , when their algorithm converted 3D animations into emerging videos ( see video above ), most of the users could recognize the animated ink-blotted figures. They also stated that when a single frame was shown to the volunteers only less than 10 percent of them could recognize the figures.

In light of the tests performed, Mitra concluded that adding motion to the ink blot images makes it easier for humans to recognize and much harder for bots to solve. Analyzing the performance of animations, as a CAPTCHA system, is still going on in his lab.

Lance Winslow in [1] mentions that if the idea of animated ink-blot CAPTCHA images works many modern interactive websites could become entirely revolutionized. Being humiliated by hackers all around the world who ruin one's online communication due to easily decipherable CAPTCHAS, the new idea has been meat with great enthusiasm and appreciation all across the world.

Some Useful Definitions :

D1 ) Cipher

In cryptography, a cipher is an algorithm for performing encryption or decryption. In non-technical usage, a "cipher" is the same thing as a "code". Firstly, the original information is known as plaintext encrypted to ciphertext. The ciphertext message contains all the information of the plain text message, but is not readable by a human or computer without the correct decrypting algorithms.

D2 ) Spam

In computing, to spam people or organizations means to send unwanted e-mails to a large number of them, usually as advertising.

Google Dictionary

D3 ) BOTS-

A "bot" is a type of malware that allows an attacker to take control over an affected computer. Also known as "Web robots", bots are usually part of a network of infected machines, known as a "botnet", which is typically made up of victim machines that stretch across the globe

Since a bot infected computer does the bidding of its master, many people refer to these victim machines as "zombies." The cybercriminals that control these bots are called botherders or botmasters.

Some botnets might have a few hundred or a couple thousand computers, but others have tens and even hundreds of thousands of zombies at their disposal. Many of these computers are infected without their owners' knowledge. Some possible warning signs? A bot might cause your computer to slow down, display mysterious messages, or even crash.

D4 ) Turing Test - Alan Turingin his 1950 paperComputing Machinery and Intelligence, proposed a test to demonstrate a machine's intelligence. The test proceeds with a judge who engages in a natural language conversation with a human and a machine, each of which tries to appear human. The two participants are placed in isolated locations. If the judge cannot tell the human and machine apart , then the machine is said to have passed the test.



Humans have the ability to aggregate information from meaningless data and to perceive something which is more useful and meaningful. This is called Emergence. This ability of humans is used by the Animated Captcha to distinguish humans from the bots.This new version of Captcha uses a mechanism to generate 3-D Objects which is difficult for any automatic algorithms to crack. Emerging Figures are used to generate Animated Captcha. Niloy Mitra,co-creator of Animated Captcha with his other colleagues has designed an algorithm by which the computer vison techniques are not able to effectively process such images.However, when a human observer is presented with an emergence image, synthesized using an object she is familiar with, the figure emerges when observed as a whole. The limit of difficulty of perceiving the image can be set by changing some parameters. Analysing the research paper presented by Niloy J. Mitra, Hung-Kuo Chu, Tong-Yee Lee, Lior Wolf, Hezy Yeshurun, Daniel Cohen-Or in ACM SIGGRAPH ASIA 2009.

(Top) This image, when stared at for a while, can reveal four instances of a familiar figure. Two of the figures are easier to detect than the others. Locally there is little meaningful information, and we perceive the figures only when observing the whole figures.

A classic example of an emergence image. Although at first sight the left image looks meaningless, suddenly we perceive the central object as the Dalmatian dog pops out.

Emergence images, when observed through small windows, look meaningless. Although we perceive the subject in the whole image, the smaller sized segments, in isolation, look like random patches. In contrast, the elephant can be recognized through similar windows of the normal shaded scene. We often fail to perceive an emergence image when the subject is in an uncommon pose. Among the users who were shown the above images, the average success rate was only 54% and 4%, respectively. When the inverted versions of these images were shown, the success rates went up to 96% and 91%, respectively.

Typical emergence images generated by our synthesis algorithm. We generate a range of examples on various subjects synthesized at different difficulty levels. Each example contains exactly one subject. (Please refer to supplementary material for other examples.

In many computer vision recognition or segmentation algorithms, the first stages comprise of multi-scale edge detection or other means of bottom-up region processing. At multiple-scales, we detect edges using standard Canny edge detector, and retain the ones that persists scales. Such curves are then linked together based on spatial proximity and curvature continuity. We observe that while on the original renderings the method successfully extracts the feature curves (right image in each box), on the emerging images the results can mostly be seen as noise. This indicates the difficulty that bottom-up algorithms face when detecting objects in the emergence images.

Emerging frog at various difficulty levels, increasing from left to right. We control the difficultly by controlling the sampling density, breaking the silhouette continuity, perturbing silhouette patches, and adding clutter using cut-perturb-paste.

Difficulty level as perceived by users and as predicted by our synthesis parameters. (Right) Perceived difficulty level in each category changes gradually. For example, 98% of the easy images were recognized by at least 80% of the observers.


This section concludes the report and recommends further work or predicts future trends on the topic.


Please follow the format below:

(1) For books:

  1. F.S. Tsai and C.K. Chan (Eds), Cyber Security, Pearson Education, 2006.
  2. (2) For articles:

  3. A.K. Jain, A. Ross, and S. Pankanti, "Biometrics: A Tool for Information Security," IEEE Transactions on Information Forensics and Security , pp. 125-143 , June 2006.
  4. (3) Internet resources:

  5. Name of author, website, date and place. (Insert where relevant).