Deep Learning Approach For Sign Language Recognition
Bambang Krismono Triwijoyo, Lalu Yuda Rahmani Karnaen, Ahmat Adil
Abstract
Sign language is a method of communication that uses hand movements between fellow people with hearing loss. Problems occur when communication between normal people with hearing disorders, because not everyone understands sign language, so the model is needed for sign language recognition. This study aims to make the model of the introduction of hand sign language using a deep learning approach. The model used is Convolutional Neural Network (CNN). This model is tested using the ASL alphabet database consisting of 27 categories, where each category consists of 3000 images or a total of 87,000 images of 200 x 200 pixels of hand signals. First is the process of resizing the image input to 32 x 32 pixels. Furthermore, separating the dataset for training and validation respectively 75% and 25%. The test results indicate that the proposed model has good performance with a value of 99% accuracy. Experiment results show that preprocessing images using background correction can improve model performance.