If you've commented anonymously on this site or thousands of others where you are not a registered member, you've come across the Completely Automated Public Turing Test to Tell Computers and Humans Apart (CAPTCHA) box - and maybe you dread it because, when it doesn't recognize the letters you think you are seeing, you are stuck.

Venu Govindaraju, a computer scientist at the University of Buffalo who pioneered machine recognition of human handwriting, says a 21st-century solution to CAPTCHA problems may rest in the early days of human culture - handwriting. 

Govindaraju and colleagues did the research in the 1990s that helped the U.S. Postal Service establish the first machines that could read handwritten addresses, a feat that many at the time, certainly in the postal worker's union, said could not be done. In 1996, after years of research, their research enabled the USPS to be able to start machine-reading of handwritten addresses, boosting efficiency and saving the agency millions of dollars each year. 

Like then, computer scientists tend to consider handwriting a hopeless cause - reading handwriting comes naturally to humans but for computers it is more complex, much like a child can recognize a cartoon picture of a chicken and say 'chicken' but computers cannot.

Govindaraju believes similar success can occur using handwriting for CAPTCHAs.
"We developed an archive that can automatically generate as many different styles of handwriting as we want," he says.  One of his grad students was recently hired by Yahoo! because of their work developing those simulated handwritten captchas.

The research is based on pattern recognition, a subfield of machine learning in computer science that is concerned with developing systems based on detecting patterns in data.

In the future, such pattern recognition could get ever bigger, like 'smart room' technologies, where indoor environments are equipped with unobtrusive cameras and microphones that can identify and track the movements and gestures of inhabitants for a broad range of applications, from providing supplemental supervision in assisted living facilities for the elderly or disabled, to monitoring office workplaces and retail establishments for security. Biometrics they are studying include hand gestures as well as the more common biometrics of facial, voice and gait recognition.

"This, too, is all pattern recognition," Govindaraju says, "but instead of letters, here, we're trying to standardize gestures.  It's like developing an alphabet of gestures so machines can be programmed to do gesture recognition. The idea is to control objects on a monitor without technology."