Google Experiments with Image Orientation CAPTCHA

After all these years, we're still trying to find the perfect CAPTCHA technique, something that will protect against automated system signups, but still be easy and not annoying for humans. Google has just released a report on what we can only hope is a step in that direction, which they call image orientation CAPTCHA.

CAPTCHA (or "Completely Automated Public Turing Test to Tell Computers and Humans Apart") is used to prevent automated computer systems from signing up at sites or posting comments. The report, called "What's Up CAPTCHA?" (.PDF) outlines a new version, which uses image orientation, forcing a user to adjust randomly rotated images to their upright orientation.

The report says:

We present a new CAPTCHA which is based on identifying an image’s upright orientation. This task requires analysis of the often complex contents of an image, a task which humans usually perform well and machines generally do not.

Given a large repository of images, such as those from a web search result, we use a suite of automated orientation detectors to prune those images that can be automatically set upright easily.

We then apply a social feedback mechanism to verify that the remaining images have a human-recognizable upright orientation. The main advantages of our CAPTCHA technique over the traditional text recognition techniques are that it is language-independent, does not require text-entry (e.g. for a mobile device), and employs another domain for CAPTCHA generation beyond character obfuscation. This CAPTCHA lends itself to rapid implementation and has an almost limitless supply of images.

We conducted extensive experiments to measure the viability of this technique.

There are currently many variations of CAPTCHA, which usually rely on distortion of a word or words. There's the regular old CAPTCHA, that makes you decipher on a single word, or ReCAPTCHA, that makes you decipher two, and even this one I wrote about years ago, KittenAuth.

However, sometimes the words are so distorted that it's difficult to get it right on the first try ... or second try ... or ... well, you get the drift. I find it extremely annoying when that happens.

According to Google, careful image selection means they won't have to distort the images, at least for now:

We ensure that our CAPTCHA can not be defeated by state-of-the-art orientation detection systems by using those systems to filter images that can be automatically recognized and oriented.

In contrast to traditional text based CAPTCHAs which introduce more noise and distortion as automated character recognition improves, we currently do not need to alter or distort the content of the images.

As advances are made in orientation detection system, these advances will be incorporated in our filters so that those images that can be automatically oriented are not presented to the user. The use of distortions may eventually be required.