APLIKASI PENDETEKSI KERUSAKAN KARDUS MENGGUNAKAN AI (ARTIFICIAL INTELLIGENCE) BERBASIS WEB

SARWAN HAMID RAMBE, NPM 2109100091 (2025) APLIKASI PENDETEKSI KERUSAKAN KARDUS MENGGUNAKAN AI (ARTIFICIAL INTELLIGENCE) BERBASIS WEB. Skripsi thesis, Universitas Labuhanbatu.

	Text COVER.pdf Download (831kB)
	Text BAB I.pdf Download (144kB)
	Text BAB II.pdf Download (817kB)
	Text BAB III.pdf Download (658kB)
	Text BAB IV.pdf Download (746kB)
	Text BAB V.pdf Download (127kB)
	Text DAFTAR PUSTAKA.pdf Download (131kB)

Abstract

Penelitian ini bertujuan untuk merancang dan membangun aplikasi pendeteksi kerusakan kardus berbasis web dengan menggunakan kecerdasan buatan (Artificial Intelligence), khususnya algoritma YOLOv8. Latar belakang penelitian ini didasari oleh tingginya tingkat kerusakan kardus dalam proses distribusi barang, yang berdampak pada kerugian industri logistik. Aplikasi ini memungkinkan pengguna mengunggah gambar kardus untuk dianalisis jenis kerusakannya, seperti sobekan, penyok, atau gesekan. Metode pengembangan sistem menggunakan pendekatan waterfall, dimulai dari analisis kebutuhan, perancangan sistem, implementasi, hingga pengujian. Dataset gambar diperoleh dari Roboflow dan model dilatih menggunakan Google Colab. Implementasi aplikasi dilakukan menggunakan bahasa pemrograman Python, framework Flask, dan basis data SQLite. Hasil penelitian menunjukkan bahwa prototipe aplikasi mampu menjalankan alur deteksi secara menyeluruh, mulai dari unggah gambar, pemrosesan oleh model AI, hingga penyajian hasil deteksi dan penyimpanan riwayat. Meskipun sistem deteksi masih berbasis simulasi karena keterbatasan server lokal, aplikasi ini telah membuktikan konsep (proof of concept) yang layak untuk dikembangkan lebih lanjut. Aplikasi ini diharapkan dapat membantu meningkatkan efisiensi dan akurasi dalam proses inspeksi kemasan di sektor logistik. Kata kunci: YOLOv8, Flask, Deteksi Kerusakan Kardus, Deep Learning, Aplikasi Berbasis Web ================================================================================================ This research aims to design and develop a web-based cardboard damage detection application using Artificial Intelligence (AI), specifically the YOLOv8 object detection algorithm. The background of this study is based on the high rate of cardboard damage during goods distribution, which can cause significant losses in the logistics industry. The application allows users to upload cardboard images, which are then analyzed to detect types of damage such as tears, dents, or scratches. The system development follows the waterfall methodology, including requirement analysis, system design, implementation, and testing. The image dataset was obtained from Roboflow and trained using Google Colab. The application was developed using Python, the Flask framework, and SQLite as the database. The results show that the prototype successfully performs the detection workflow, from uploading images to presenting detection results and storing detection history. Although the detection backend is still simulated due to local server limitations, the system demonstrates a strong proof of concept for further development. This application is expected to enhance efficiency and accuracy in the packaging inspection process within the logistics sector. Keywords: YOLOv8, Flask, Cardboard Damage Detection, Deep Learning, Web-Based Application

Item Type:	Thesis (Skripsi)
Uncontrolled Keywords:	YOLOv8, Flask, Deteksi Kerusakan Kardus, Deep Learning, Aplikasi Berbasis Web===============YOLOv8, Flask, Cardboard Damage Detection, Deep Learning, Web-Based Application
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General) Z Bibliography. Library Science. Information Resources > Z665 Library Science. Information Science Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4050 Electronic information resources Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4450 Databases
Divisions:	Fakultas Sains Dan Teknologi > Sistem Informasi
Depositing User:	Unnamed user with email repository@ulb.ac.id
Date Deposited:	29 Oct 2025 07:17
Last Modified:	29 Oct 2025 07:17
URI:	http://repository.ulb.ac.id/id/eprint/1882

Actions (login required)

View Item