IT Wizard is happy to share with you the release of Mezzofanti, an Android phone application that allows reality augmentation (as seen in camera mode) with information about any text in view. The augmentation is done through an internal text-recognition engine.
The application at a glance:
1. Select a text of interest (ex: menu at a restaurant, tourist information sign, etc), and take a photo. The internal-OCR engine will start text-recognition.
2. Use Mezzofanti to translate the text (or parts of it) to any one of over forty languages, or to google it on the web, find more information in the dictionary or on Wikipedia.
Technologies:
- the internal-OCR engine is a modified version of Tesseract 2.03, currently developed by Google under Apache License v2.0. - The engine is particularly optimized for 5 languages (English, German, Spanish, Italian, and French), but it works fairly well with any other language that uses the Latin alphabet.
- the interface between the OCR engine (C++) and Java is done in JNI
- the main core is programmed in Java
- the translation engine is based on Google-translate
License information:
- all Mezzofanti code is available for free under the Apache License v2.0 and it is posted at google code.
On our website http://www.itwizard.ro you can also find:
- a tutorial about Mezzofanti's features and internal architecture
- all the Mezzofanti code, and a tutorial on how to install & run it from your own SDK
- how to interface the Tesseract OCR with your own code, JNI/NDK how-to
- many other tutorials and examples for Android
We encourage dialog, so please leave us some feedback. We're particularly interested in collaboration opportunities on new and exciting projects for mobile platforms.
Looking forward for your thoughts! ;)
Enjoy using Mezzofanti,
IT Wizard
