Sign In

Communications of the ACM

ACM Careers

Optical Character Recognition for Graffiti

View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
graffiti in Beijing

Credit: Los Angeles Review of Books

Researchers in China have recognized that optical character recognition (OCR) has matured and can identify and extract information from documents that use standard writing styles. Now a team has developed an algorithm that can, with fine granularity, extract information from what might be loosely terms graffiti, convoluted handwriting that might even be indecipherable to a human reader.

Jiashuang Xu and Zhangjie Fu of the Computer and Software College at Nanjing University of Information Science and Technology, and Xingyue Du of the School of Humanities and Social Sciences at Xi'an Polytechnic University, provide details of their approach in "Graffiti-Writing Recognition With Fine-Grained Information," published in the International Journal of Computational Science and Engineering.

So far the team has trained their system to recognize 26 letters of the English alphabet with 85.96 percent accuracy and are now working on extending and improving the technology. The system utilizes a motion-detection approach rather than requiring touch input and so could be adapted for non-screen input devices such as wearables, where one might gesture to a device embedded in clothing, for instance.

From Tech Xplore
View Full Article


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account