Implementasi Image Processing dan Algoritma Deep Learning Untuk Klasifikasi Sampah Plastik

Putri, Frenti Haryatama (2024) Implementasi Image Processing dan Algoritma Deep Learning Untuk Klasifikasi Sampah Plastik. S1 thesis, Universitas Jambi.

[img] Text
ABSTRAK.pdf

Download (2MB)
[img] Text
BAB I.pdf

Download (11MB)
[img] Text
BAB V.pdf

Download (1MB)
[img] Text
COVER.pdf

Download (939kB)
[img] Text
DAFTAR PUSTAKA.pdf

Download (6MB)
[img] Text
FULL SKRIPSI.pdf
Restricted to Repository staff only

Download (4MB)
[img] Text
HALAMAN PENGESAHAN.pdf

Download (435kB)
Official URL: https://repository.unja.ac.id/

Abstract

Aktifitas manusia tidak dapat dipisahkan dengan kegiatan produksi dan konsumsi yang berdampak pada timbulnya sampah seperti penggunaan plastik. Oleh karena itu, deteksi dan pemilahan sampah sebaiknya dilakukan pada tahap awal pengelolaan sampah untuk memaksimalkan jumlah sampah yang dapat didaur ulang. Penelitian ini bertujuan untuk menerapkan image processing dan algoritma deep learning dalam klasifikasi sampah plastik, serta menguji performa sistem klasifikasi tersebut. Metode penelitian yang digunakan mengacu pada tahapan penelitian yaitu studi pustaka, pengumpulan data, pre-processing, perancangan sistem, implementasi, pengujian, evaluasi dan analisis data. Hasil penelitian menunjukkan bahwa image processing dan deep learning dapat diimplementasikan untuk klasifikasi sampah plastik dengan berbagai kelas, menggunakan model YOLOv8. Sistem klasifikasi sampah plastik memperoleh akurasi, presisi, recall, dan F1 score yang tinggi, yaitu 98,7%, 1, 0,98, dan 0,99. Kata Kunci: Klasifikasi, Deep Learning, Pengolahan Citra, YOLO Human activities cannot be separated from production and consumption activities which have an impact on the generation of waste, such as the use of plastic. Therefore, waste detection and sorting should be carried out at the initial stage of waste management to maximize the amount of waste that can be recycled. This research aims to apply image processing and deep learning algorithms in plastic waste classification, as well as testing the performance of the classification system. The research method used refers to the research stages, namely literature study, data collection, pre-processing, system design, implementation, testing, evaluation and data analysis. The research results show that image processing and deep learning can be implemented for the classification of plastic waste with various classes, using the YOLOv8 model. The plastic waste classification system obtained high accuracy, precision, recall and F1 scores, namely 98.7%, 1, 0.98 and 0.99. Keywords: Classification, Deep Learning, Image Processing, YOLO

Type: Thesis (S1)
Uncontrolled Keywords: Klasifikasi, Deep Learning, Pengolahan Citra, YOLO
Subjects: L Education > L Education (General)
Divisions: Fakultas Teknik > Teknik Elektro
Depositing User: PUTRI
Date Deposited: 06 Aug 2024 06:51
Last Modified: 06 Aug 2024 06:51
URI: https://repository.unja.ac.id/id/eprint/69839

Actions (login required)

View Item View Item