Implementasi Metode Yolo V8 dalam Identifikasi Tajwid: Hukum Bacaan Nun Mati pada Al-Quran Secara Realtime Berbasis Android

Abstract

The difficulties of Muslims in knowing the science of tajweed, especially the law of reading Nun Mati, is a challenge for most Muslims when reading the Qur'an. Traditional learning media are often perceived as less effective and boring, so innovation in digitising learning is needed. This research aims to implement the You Only Look Once version 8 (YOLOv8) algorithm to create an Android-based application that can identify the reading law of Nun Mati in real time. The YOLOv8 method is used because of its advantages in detecting objects with high accuracy and processing efficiency. The results showed that the system was able to detect 28 types of Nun Mati tajweed laws with an accuracy rate of 97,7%. In addition, the developed model achieved an average precision of 100%, a recall of 93,8% and an F1 score of 96,2%. The system can operate in real time on Android devices with excellent performance, despite obstacles in low light conditions. In conclusion, the YOLOv8 algorithm proved to be effective in detecting the Nun Mati recitation law, and the developed application has great potential to help Muslims improve the fluency and accuracy of Qur'an recitation. This research also shows that the YOLOv8-based system has better performance compared to models from previous studies for the same object.