Logo Lanfrica

AchrafBir/Reconnaissance-et-Traduction-Automatique-de-la-Langue-des-Signes-Marocaine

Domain:

natural language processing

Record type:

software
Creator:
Ach
Host:
Reconnaissance et Traduction Automatique de la Langue des Signes Marocaine : Une Approche par Vision par Ordinateur et Stockage MongoDB # Moroccan Sign Language Recognition Project ## Overview This project implements a Convolutional Neural Network (CNN) to recognize Moroccan Sign Language (MSL) signs. It includes data loading, preprocessing, model training, evaluation, and storage of results and misclassification analysis in MongoDB. Additionally, it provides a real-time sign detection script using OpenCV. ## Project Structure * **morrocan_sign_language.ipynb:** Jupyter Notebook containing the main code for data loading, model creation, training, evaluation, and analysis. * **moroccan_sign_language_model.h5:** Saved Keras model file (created by the notebook). * **README.md:** This file. ## Technologies Used * **Python 3:** Programming language * **TensorFlow/Keras:** Deep learning framework * **OpenCV:** Computer vision library for image processing * **NumPy:** Numerical computing library * **Pandas:** Data analysis library * **Matplotlib & Seaborn:** Data visualization libraries * **Scikit-learn:** Machine learning library * **PyMongo:** MongoDB driver for Python * **MongoDB Atlas:** Cloud database service (or local MongoDB instance) ## Setup and Installation ### 1. Prerequisites * **Python 3.x:** Make sure you have Python 3 installed. * **MongoDB Atlas Account (or local MongoDB instance):** You will need a MongoDB Atlas account to store the data, or a local instance of MongoDB running. * **Kaggle Account:** To access the dataset directly from Kaggle. ### 2. Install Python Packages ```bash pip install pymongo dnspython tensorflow scikit-learn opencv-python pandas matplotlib seaborn ``` ### 3. MongoDB Configuration #### MongoDB Atlas Connection String: * **1. Create a MongoDB Atlas cluster :** Choose a free tier cluster for testing purposes. * **2. Create a database user :** Make sure to grant the user read and write access to your chosen database. * **3. Whitelist your IP address :** You need to allow ur IP address in the Network Access section to allow co …

Licenses

Similar