This project classifies Indian Sign Language (ISL) gestures, including both digits and alphabets, using image data and machine learning for accurate gesture recognition. The main goal is to enable real-time gesture classification, supporting applications such as assistive communication technologies for the hearing impaired.

Key Features: Gesture Classification: Classifies hand gestures into digits and alphabets. Feature Extraction: Uses image data and machine learning for feature extraction and classification. High Accuracy: Achieved high accuracy in classifying gestures from the dataset. Real-time Deployment: Trained model can be deployed for real-time gesture recognition on embedded devices.

Dataset: The dataset includes images of hand gestures representing digits (0-9) and alphabets (A-Z) in Indian Sign Language. Each gesture category contains 1,200 images. The images were processed, augmented, and labeled to ensure efficient training.

Tools and Libraries Used: Edge Impulse Studio: Used for designing the impulse, training the model, and deploying it for gesture recognition. Image Data: Hand gestures representing digits and alphabets in Indian Sign Language. Edge Impulse’s Built-in Tools: Utilized for feature extraction, model training, and evaluation within the platform.

The trained model can be deployed on Edge Impulse-supported devices for real-time gesture recognition, enabling applications in assistive communication for the hearing impaired.