<![CDATA[For deaf or hard of hearing people, sign language is a primary means of communication, but low public understanding makes social engagement difficult. Researchers now use computer vision technology and Convolutional Neural Network (CNN) to detect sign language movements. Problems such as overfitting and missing gradients still exist. Using CNN and ResNet-34 architecture, as well as image augmentation to overcome this problem, this research builds a deep learning-based sign language detection model. The Indonesian Sign Language System (SIBI) dataset was used to test the model. The test results show that the model with image augmentation trained for more than 50 epochs obtained an accuracy of 99.4%, precision of 99.5%, recall of 99.5%, and an F1 score of 99.5%. The model without image augmentation produced an accuracy of 99.4%, recall of 99.3%, F1 score of 99.3%, and precision of 99.4%. ResNet-34 architecture overcomes the problem of missing gradients, while image augmentation avoids overfitting and improves model accuracy.]]>
