Overview / Usage

The major goal of this project is to translate sign language to speech and text in real time. It provides a helping hand to speech impaired and disabled persons to communicate with rest of the world.

Methodology / Approach

Firstly, the image is being processed to the system through the camera. Then grayscale conversion of the image is done for simplicity and accuracy. The contours are being detected and convex hull are formed within the image. Currently, the project uses a predefined set of information in accordance with the ASL standards .It calculates the angle between edges and and area of hull and matches with the predefined standards. And provides the best outcome accordingly.
But the under development approach is to train the model to the best possible extent so that it provides the most accurate output and ignore the noise within the image.
For the output, currently the screen and the audio speakers are used , which display or play when the result matches the data sets.

Technologies Used

1.Open CV
2. Python Language
3. Edge Detection
4. Speakers and Screen
5. Camera