Indian Institute of Technology Delhi
"Serving individuals with special needs through research and innovation in affordable assistive technologies"
You are Here : Home > Research > Student Project



Audio streaming of Projection for the visually impaired
The aim of this project is to enable visually impaired people to understand the contents of a presentation displayed using a projector on a screen. The device, target of this project, will be able to convert the content displayed on the screen and transfer it to the user's laptop as a audio stream.

Shubham Solanki
Language Recognition and Text Extraction in Road Sign Boards
To assist in outdoor navigation by creating an automated system to identify and classify road sign board images based on language and extracting the text. Description : The project consists of three phases Phase 1 : Image Pre-Processing & Text Recognition Detecting text in natural scene images, as opposed to scanned pages of documents is challenging due to high variability of text appearance and different languages. This phase therefore, focuses segmentation of foreground text and complex background from natural scene images of sign boards. The input image to the system will be a camera captured image. Phase 2 : Language Recognition (Identification of the language in the image obtained from phase 1) This is the most important phase of the project. The basic problem involved here is the classification of the textual regions of the image(obtained from phase 1) into different classes (of languages) using image based features. It involves training the classifier by applying a suitable learning algorithm. Phase 3 : Text Extraction After identifying the language, the image is fed to an OCR engine specific to that language for extracting the text.

B.Sai Deep, Akshay Suresh Bhat