Implementasi metode You Only Look Once v11 untuk deteksi objek: Studi kasus hukum tajwid pada Al-Quran

Halizah, Nur (2025) Implementasi metode You Only Look Once v11 untuk deteksi objek: Studi kasus hukum tajwid pada Al-Quran. Sarjana thesis, UIN Sunan Gunung Djati Bandung.

[img]
Preview
Text
1-cover.pdf

Download (257kB) | Preview
[img]
Preview
Text
2-abstrak.pdf

Download (287kB) | Preview
[img]
Preview
Text
lembarpernyataankaryasendiri.pdf

Download (425kB) | Preview
[img]
Preview
Text
3-daftarisi.pdf

Download (246kB) | Preview
[img]
Preview
Text
4-bab1.pdf

Download (315kB) | Preview
[img] Text
5-bab2.pdf
Restricted to Registered users only

Download (988kB) | Request a copy
[img] Text
6-bab3.pdf
Restricted to Registered users only

Download (1MB) | Request a copy
[img] Text
7-bab4.pdf
Restricted to Registered users only

Download (3MB) | Request a copy
[img] Text
8-bab5.pdf
Restricted to Registered users only

Download (280kB) | Request a copy
[img] Text
9-daftarpustaka.pdf
Restricted to Registered users only

Download (254kB) | Request a copy
[img] Text
10-lampiran.pdf
Restricted to Repository staff only

Download (320kB) | Request a copy

Abstract

Membaca Al-Quran dengan baik adalah suatu keharusan bagi umat Muslim, sehingga mengenali dan memahami setiap hukum tajwid berperan penting untuk memastikan bacaan Al-Quran sesuai dengan kaidahnya. Akan tetapi, hukum tajwid terdiri dari berbagai jenis, yang menyebabkan tidak jarang pembaca Al-Quran mengalami kesulitan dalam mengenali dan memahami hukum tajwid dengan tepat. Sebagai upaya dalam mengatasi tantangan tersebut, penelitian ini bertujuan untuk membangun model deteksi objek pada Al-Quran menggunakan metode YOLO versi terbaru, yaitu YOLOv11. Sedangkan metode penelitian yang digunakan mengikuti pendekatan CRISP-DM yang mencangkup business understanding, data understanding, data preparation, modelling, evaluasi dan deployment. Model dilatih menggunakan dataset gambar per halaman, per ayat, dan per kata pada AlQuran dengan cetakan Mushaf Standar Indonesia (MSI) dan Al-Quran cetakan Madinah. Dataset gambar yang terkumpul sebanyak 2.000 dilakukan proses labelling berdasarkan jenis-jenis tajwidnya dan dilakukan augmentasi data menggunakan platform Roboflow. Hasil penelitian ini menunjukkan bahwa model YOLOv11 mampu mendeteksi dan memprediksi 34 jenis tajwid dengan baik, nilai precision yang dihasilkan sebesar 0.683, nilai recall sebesar 0.75, nilai mAP50 sebesar 0.739, dan nilai mAP50-95 sebesar 0.541. Selain itu, model yang dibangun diimplementasikan pada sebuah sistem berbasis web menggunakan framework Flask. Sistem dapat digunakan dengan mengunggah gambar, dan hasil deteksi akan keluar secara otomatis pada halaman web. Diharapkan sistem ini dapat menjadi media belajar dalam mempelajari dan mengenali hukum tajwid pada Al-Quran. Reading the Quran properly is a must for Muslims, so recognizing and understanding each law of tajweed plays an important role in ensuring that the reading of the Quran is in accordance with its rules. However, the laws of tajweed consist of various types, which often causes readers of the Quran to have difficulty in recognizing and understanding the laws of tajweed correctly. As an effort to overcome these challenges, this study aims to build an object detection model in the Quran using the latest version of the YOLO method, YOLOv11. The research method used follows the CRISP-DM approach which includes business understanding, data understanding, data preparation, modeling, evaluation and deployment. The model is trained using image datasets per page, per verse, and per word in the Quran with the Indonesian Standard Mushaf (MSI) printing and the Medina printed Quran. The collected image dataset of 2,000 was labeled based on the types of tajweed and data augmentation was carried out using the Roboflow platform. The results of this study indicate that the YOLOv11 model is able to detect and predict 34 types of tajweed well, the resulting precision value is 0.683, the recall value is 0.75, the mAP50 value is 0.739, and the mAP50-95 value is 0.541. In addition, the model built is implemented on a web-based system using the Flask framework. The system can be used by uploading images, and the detection results will appear automatically on the web page. It is hoped that this system can be a learning medium in studying and recognizing the laws of tajweed in the Qur'an.

Item Type: Thesis (Sarjana)
Uncontrolled Keywords: Al-Quran; Tajweed Law; Object Detection; YOLOv11; Flask
Subjects: Systems > Computer Modeling and Simulation
Al-Qur'an (Al Qur'an, Alquran, Quran) dan Ilmu yang Berkaitan
Divisions: Fakultas Sains dan Teknologi > Program Studi Teknik Informatika
Depositing User: Nur Halizah
Date Deposited: 19 Aug 2025 02:22
Last Modified: 19 Aug 2025 08:46
URI: https://digilib.uinsgd.ac.id/id/eprint/115284

Actions (login required)

View Item View Item