What's up CAPTCHA?

摘要: We present a new CAPTCHA which is based on identifying an image's upright orientation. This task requires analysis of the often complex contents image, humans usually perform well and machines generally do not. Given large repository images, such as those from web search result, we use suite automated orientation detectors to prune images that can be automatically set easily. then apply social feedback mechanism verify remaining have human-recognizable The main advantages our technique over traditional text recognition techniques are it language-independent, does not require text-entry (e.g. for mobile device), employs another domain generation beyond character obfuscation. lends itself rapid implementation has almost limitless supply images. conducted extensive experiments measure viability this technique.