Rosetta Phone

Unable to read or interpret written text can pose major problems for people in many circumstances, and Rosetta Phone is a system that will aid in interpreting this written text. The objective of this research project is to develop a complete system investigating both hardware platforms and software programming environments capable of real-time, autonomous, visual interpretation of written text. In Phase I, the proof-of-concept system developed shows that the Rosetta Phone concept is achievable using Handheld Mobile Devices (HMDs) available on the market today. Using open standards we have developed a system which will run not only on HMDs, but also on any computer running Linux or other UNIX-like operating systems, without making any modifications to the code. All the tools developed in Phase I are able to run on the device, and are able to process an image and generate the desired output. Phase II will focus on improving the software, expanding the database of characters and languages, and continuing to research commercially-available and militarized HMDs to ensure sufficient camera resolution, data storage, memory, and processing capabilities. Improvements to the software will include image fidelity, processing, and additional language support. Rosetta Phone could have military applications in foreign countries where translation of signs or other text could provide vital information to the warfighter. Other obvious applications are a reading assistant for the blind or as a portable translation device for tourists to foreign countries.