The project was done as a coursework during CS course in HSE.
The project is based on Image Processing techiques as Feature Extraction and Contour Analysis. The recognition is hardcoded and calibrated according to my laptop webcam.
As a result the program is capable of recognizing about 15 gestures of RSL alphabet:
(А, Б, В, Е, Ж, И, Н, О, С, У, Ф, Ш, Ь, Ы, Ю)
WARNING! The program is extremely sensitive to the environment. Recommended environment is a clear wall as a background and a medium level of lamp light.
The program was tested on Ubuntu 16 and Windows 10.
Dependencies:
- OpenCV 3.2.0
- Python 2.7
- NumPy
- Matplotlib