Page 436 -
P. 436

FIGURE 14.1
                   reCAPTCHA: (A) The original reCAPTCHA asked users to type in words that could
                   not be recognized via optical-character recognition. A sample of text that might have
                   been recognized is shown, along with a depiction of how the text might be distorted
                   before being presented to the user. (B) More recent reCAPTCHA tasks involve image
                   classification, such as choosing images from a set that match a specified criteria—in
                   this case, images containing fireworks.
                    (A) From von Ahn, L., Maurer, B., McMillen, C., Abraham, D., Blum, M., 2008. ReCAPTCHA: humanbased
                                      character recognition via web security measures. Science 321, 1465–1468;
                                   (B) From https://www.google.com/recaptcha/api2/demo (accessed April 8, 2017).

                                                                           (Continued)
   431   432   433   434   435   436   437   438   439   440   441