Captcha comes from "Completely Automated Public Turing test to
tell Computers and Humans Apart". It is used by web sites to
distinguish human web users from robot spammers.
Recently, in just minutes, an artificially intelligent machine
cracked ond of those jumbled text sequences. The smart machine
can be trained in a matter of minutes using just a few hundred
example characters, researchers said. It works with multiple
different styles of captcha and can also be re-purposed to
identify handwritten digits, recognize text in photos of
real-world scenarios, and detect non-text objects in images.
That is because the company designed the system to mimic the way
the brain identifies objects after seeing just a few examples
and recognize them in strange new configurations.
Text-based captchas as above work, because unlike humans,
computers struggle to recognize the distorted and partially
hidden characters that make them up. Though many current
machine-learning systems that can solve them exist, they must be
trained on millions of images to work.
Many websites have
moved away from text-based captchas, and are using image-based
tests and data on mouse movements, or cookies to analyze whether
users are human or machine. Some captchas are also difficult for
humans to accurately decipher on first try.
No comments:
Post a Comment
Comments