Assistive Technology

 

Skill Level	Area of Focus	Operating System	Platform/Hardware
Intermediate	Computer Vision, Education	Linux	DragonBoard 410c

The project uses a camera input connected to the DragonBoard™ 410c from Arrow Electronics, and is designed to identify and read text in an image, and either recognizes it in its native language or translates it to another language.

Objective

The main goal of the project is to use the power of the DragonBoard 410c to help people with reading difficulty by running a computer vision application for text detection and recognition. Using powerful libraries, this app can find text in the image captured by the camera, transform it into a text string, and read that string using voice emulation.

Materials Required / Parts List / Tools

• DragonBoard 410c boards
• 12V wall adapter
• Micro SD Card (16GB)
• Logitech C920 Webcam
• USB Keyboard
• HDMI Monitor

Source Code / Source Examples / Application Executable

Source code and tutorial on GitLab

Build / Assembly Instructions

The build instructions can be found in the README file in the GitLab link. Basically, you have to build the libs and install the application on the board.

Usage Instructions

You can use the project as a text to speech app, or you can use it for translation. Point the camera at the text and select the language for translation. The app will read it in your native language. A usage guide for the app interface can be found in the GitLab repository as well.