Sign Language Translation System
Keywords:
Sign Language Detection, Deep Learning, Convolutional Neural Network (CNN), Detection Transformer (DETR), Hand Gesture Recognition, Computer Vision, Real-Time Detection, Image Processing, Assistive Technology, Text ConversionAbstract
Sign language serves as a vital communication medium for individuals with hearing and speech impairments.
However, communication between sign language users and non-users remains a major challenge. This project
proposes a real-time sign language detection system that utilizes deep learning and transformer-based object
detection techniques to recognize hand gestures and convert them into meaningful text.
The system captures live video through a webcam and performs hand detection and preprocessing to extract the
region of interest. Advanced deep learning models, such as Convolutional Neural Networks (CNNs) and Detection
Transformers (DETR), are employed to accurately classify hand gestures. Each video frame is processed in real
time, and the corresponding sign is predicted and displayed as text on the screen.
To improve recognition accuracy, preprocessing techniques including image normalization, resizing, and noise
reduction are applied. The model is trained using labeled datasets containing various hand gestures, ensuring
robust and reliable performance under different conditions. The proposed system provides efficient, fast, and
accurate gesture recognition suitable for real-world applications.
This project helps bridge the communication gap between sign language users and the wider community by
enabling seamless interaction. Furthermore, the system can be extended to support sentence formation, speech
conversion, and integration with assistive technologies for enhanced accessibility.
