Sign Language Detection System Using CNN and HOG: Bridging the Communication Gap for Deaf and Hearing Communities

In order to effectively extract features, this research proposes a novel real-time sign language detection system that combines Convolutional Neural Networks (CNN) with Histogram of Orientated Gradients (HOG). The device bridges the communication gap between the hearing and deaf communities by successfully recognizing both static and dynamic gestures in real time using a typical web camera. This technology, which is intended to be both economical and easy to use, removes the need for pricey hardware, allowing for wider application. Testing demonstrates that the system can accurately and consistently identify a large variety of hand gestures, increasing its applicability for everyday use in a variety of contexts. The technology has potential applications in public areas, educational institutions, and businesses where effective communication between sign language users and non-users is crucial since it fosters social inclusion.