Thaindian News

Washington, Sep 10 (IANS) Computers can copy many aspects of human behaviour, but they don’t yet possess our ability to recognise distorted images, says a team of researchers.
“Our goal is to seek a better understanding of the fundamental differences between humans and machines and utilise this in developing automated methods for distinguishing humans and robotic programmes,” said co-author James Z. Wang.

Wang is an associate professor in Penn State University (PSU) College of Information Sciences and Technology.

Wang, along with Ritendra Datta, a PSU doctorate and Jia Li, associate professor of statistics at Penn State, explored the difference in human and machine recognition of visual concepts under various image distortions.

Researchers used those differences to design image-based CAPTCHAs (Completely Automated Public Turing Test to Tell Computers and Humans Apart), visual devices used to prevent automated network attacks.

Many e-commerce websites use CAPTCHAs, which are randomly generated sets of words that a user types in a box provided in order to complete a registration or purchasing process. This is done to verify that the user is human and not a robotic programme.

Although the scope of the human users was limited, the results proved that robotic programmes were not able to recognise distorted images.

In other words, a computer recognition programme had to rely on an accurate picture, while humans were able to tell what the picture was even though it was distorted.

Wang said he hopes to work with developers in the future to make IMAGINATION a CAPTCHA programme that websites can use to strengthen the prevention of automated network attacks.

Even though machine recognisability does not exceed human recognisability at this time, Wang says that there is a possibility that it will in the future, says a PSU release.

These findings appeared in the September issue of IEEE Transactions on Information Forensics and Security.