CAPTCHA test designed to identify human cracked by the machinecheat or take advantage of in order to prevent users from malicious software attacks, many Nike SB sites and applications use CAPTCHA (Completely Automated Public Turing Test to Tell Computers and Humans Apart, automatic distinction between computer and human Turing test) to verify whether the logged on user at the other end Humanity. Now, however, this approach has been ineffective, because there is a piece of software capable of this test. This shows that close to human direction toward artificial intelligence and a step forward. Allowing users to recognize distorted text often is the site to determine whether other human means. In theory, intended to distinguish between humans and machines CAPTCHA can have many forms, but the distorted images of text form proved to be an effective way to prevent malware and SPAM in the robot program. This is because when writing in a twisted, overlapping or random lines, dots and color images covering the form of software is difficult to decipher. And in this respect mankind has superior ability, often only need Kanji Yan can recognize various changes in form of text. But a startup called Vicarious this week nike dunk heels announced that it has developed an algorithm that can beat the text-based CAPTCHA. Which is considered the most difficult of Google's reCAPTCHA test its recognition rate of 90%, while in Yahoo, Paypal and CAPTCHA.com performance even better. The company beat the virtual neurons CAPTCHA tool is a virtual neurons. These virtual neurons are access to on a network modeled on the human Nike Air Force 1 brain. Each node of the network (virtual neurons) clear division of labor. First there will be the first layer of input nodes to detect the real world, say judge a particular pixel in the image is white or black. On the basis of the first two-layer node detects whether the existence of a special arrangement Air Jordan 11 of pixels. If Nike Shoes present, to the layer 3 nodes to determine whether the layout arrangement of a fully or Air Jordan Others partially matching shape. This process will be repeated between the third layer and 8 layers, up to a maximum signal transfer between eight million nodes. Finally, the network will give the image they think is most likely to contain the text. The traditional image recognition method is generally stored in a particular character images into the database, then the image will be recognized with which seized control record. But this identification process neural network is not mechanically retrieved control, but with the human identification process is similar. First, there is a neural network training process, learning those that have been identified in the CAPTCHA code and activities letters video, and then learn and continually improve their awareness. Although the neural network-based systems have been widely deployed, but because the company did not disclose further details, so the algorithm is unknown count on technological breakthroughs. But if indeed break the text-based CAPTCHA, the implications would be very significant, because the recognized character requires very sophisticated understanding. Currently Vicarious not yet intend to launch the product, but plan to challenge more of the Turing test. Its goal is to make the machine under a complex scene to say something happened, or to learn how to adapt to an easy Air Jordan 7 Retro job can be performed elsewhere. Such intelligence can be applied to things like robot butler above, in order to handle all kinds of situations in the clutter of the human environment. Vicarious's CEO Scott Phoenix say their focus is to solve the fundamental problem, move to crack CAPTCHA belong unintentional. Unless noted, articles are original or compiled site, please indicate: articles from 36 Krypton