/Sign-Language-Detector

A python based app which can convert the hand gesture to corresponding text label in real time.

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

banner

A python based app which can convert the hand gesture to corresponding text label in real time.

GitHub repo size    License: GPL v3

Basic Overview

" Talk to a man in a language he understands, that goes to his head. Talk to him in his own language, that goes to his heart. "

This project introduces Sign Language recognition system which can recognize American Sign Language.


The project makes use of Convolutional Neural Network (CNN) Algorithm for training and to classify the images. The proposed system was able to recognize 10 American Sign gesture alphabets with high accuracy.

Model has achieved a remarkable accuracy of 97.89%

  

Getting Started

It requires python version 3.6 or later as to synchronize with tensorflow.

  • collect_data.py file will help in creating your own dataset using webcam.

  • cnn_model.py file will use Convolutional Neural Network (CNN) to train the model and store it in the form of hadoop distributed (h5) format.

  • gesture_predict.py file will recognise the gesture as per the trained dataset.

  • SignDetector.py file contains the code used to built the UI of the application.

Pre-requisites

  • python 3.6 or later
  • tkinter module
  • tensorflow
  • keras
  • opencv
  • pillow
  • numpy
  • imutils
  • matplotlib

Installations

1. Install python 3

Python is usually installed by default on most modern systems. To check what python version you currently have, open a terminal and run the following command:

python --version

This should output some information on the installed Python version. You can also install python by following these instructions: https://installpython3.com/

2. Install tensorflow framework

Install tensorflow framework with the following command

python -m pip install tensorflow --trusted-host files.pythonhosted.org --trusted-host pypi.org --trusted-host pypi.python.org

For upgradation use the following command

pip install tensorflow==2.0.0-alpha0

3. Install keras

Install keras with the following command

pip install keras

Note: if keras doesnt work just replace keras.model to tensorflow.keras.model and keras.preprocessing to tensorflow.keras.preprocessing

4. Install OpenCV

Install opencv for python with the following commands

pip install opencv-python==3.4.2.16

pip install opencv-contrib-python==3.4.2.16

5. Install pillow

Install PIL with the following command

pip install pillow

6. Install numpy

Install numpy with the following command

pip install numpy

7. Install imutils

Install imutils with the following command

pip install imutils

8. Install matplotlib

Install matplotlib with the following command

pip install matplotlib

How to run the project ?

There are 2 ways to run the project which are as follows:

  1. You will be able to run the project by simply running the SignDetector.py file.

  2. You will be able to execute the app by running the following command from terminal in the project directory.

On Windows:

.\SignDetector.bat

On Unix:

./SignDetector.sh

Typing SVG