For people with hearing loss or for those using sign language (hand gestures) as their major form of communication, a sign language recognition system is used. The major aim of sign recognition utilizing hand gestures is to allow successful communication and accessibility. It was investigated which algorithms best distinguish a hand motion. The diagonal sum algorithm was discovered to have the highest accuracy rate. A custom algorithm eliminates each training gesture's background during the preprocessing stage. The image is then transformed into a binary image, and the sums of each diagonal element in the image are calculated. The process includes machine learning techniques particularly deep learning architectures like Convolutional Recurrent Neural Network (RNN) and Convolutional Neural Network (CNN), which are used to specify and interpret these gestures from images. Accuracy rates exceeding 95% are commonly achieved for well-defined and relatively basic hand motions, such as recognizing a restricted range of static signs, especially with well-curated datasets and powerful machine learning algorithms. Gesture recognition systems created for certain controlled situations, such as isolated hand motions under perfect lighting, may achieve even better accuracies. However, getting high accuracy gets increasingly difficult as the complexity of gestures grows, such as recognizing dynamic or continuous hand movements and gestures with fluctuations in illumination, backdrop, and user behavior.
Performance Evaluation of Sign Recognition Protocol Using Hand Gesture Mechanism
2023-11-22
372936 byte
Conference paper
Electronic Resource
English
Real-Time Hand Gesture Recognition for Robot Hand Interface
British Library Conference Proceedings | 2014
|Deaf Sign Language using Automatic Hand Gesture Robot Based on Microcontroller System
BASE | 2019
|