Reading braille requires sensitivity of the hand, memorizing every combination of dots that are formed. Getting to know Braille takes a relatively long time. The aim of this research is to understand Braille letters using an Android-based smartphone.The research method used is descriptive quantitative using the Java, Kotlin, Python programming languages, and the Convolutional Neural Network (CNN) algorithm with the MobilenetV2 architectural model. The results of testing the CNN algorithm with the MobilenetV2 architecture model have an accuracy value of 0.9183. This shows that the application detection performance is less than optimal.
Copyrights © 2024