Rancang Bangun Sistem Deteksi Dan Klasifikasi Sampah Otomatis Menggunakan Convolutional Neural Network Untuk Pengenalan Sampah Organik, Anorganik, Dan B3

Sudarnata, I Made Ricky (2025) Rancang Bangun Sistem Deteksi Dan Klasifikasi Sampah Otomatis Menggunakan Convolutional Neural Network Untuk Pengenalan Sampah Organik, Anorganik, Dan B3. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2040211030-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (6MB) | Request a copy

Abstract

Pengelolaan sampah di lingkungan Departemen Teknik Elektro Otomasi (DTEO) masih menghadapi kendala, khususnya pada proses pemilahan sampah yang belum berjalan secara efisien. Kurangnya sistem otomatis yang mampu membedakan jenis sampah secara real-time menjadi salah satu faktor yang menghambat pengelolaan yang terstruktur. Penelitian ini bertujuan untuk mengembangkan sistem klasifikasi dan pemilahan sampah otomatis berbasis Convolutional Neural Network (CNN) dengan algoritma YOLOv8. Sistem ini dirancang untuk mengenali tiga kategori utama sampah, yaitu organik, anorganik, dan bahan berbahaya dan Beracun (B3), dari citra yang ditangkap oleh kamera. Model YOLOv8n dilatih menggunakan 1000 gambar yang terbagi ke dalam sepuluh kelas objek. Hasil klasifikasi digunakan untuk mengontrol servo motor melalui mikrokontroler ESP32 agar sampah diarahkan ke tempat pembuangan yang sesuai. Sistem juga dilengkapi sensor load cell untuk mengukur berat dan sensor ultrasonik untuk mengukur volume sampah. Seluruh data klasifikasi dan hasil pembacaan sensor ditampilkan secara real-time melalui OLED dan dashboard web berbasis MySQL. Pengujian menunjukkan bahwa model mampu mencapai precision sebesar 1,00, dengan nilai recall dan F1-score berkisar antara 0,80 hingga 0,95. Sensor load cell menunjukkan akurasi tinggi dengan error maksimal sebesar 0,20%, sedangkan sensor ultrasonik mencatatkan hasil deteksi volume tanpa error. Hasil ini menunjukkan bahwa sistem mampu melakukan klasifikasi dan pemilahan sampah secara otomatis, akurat, dan responsif, serta berpotensi diterapkan untuk mendukung pengelolaan sampah yang lebih efektif di lingkungan kampus.
=================================================================================================================================
Waste management at the Department of Electrical Automation Engineering (DTEO) still faces several challenges, particularly in the sorting process, which has yet to operate efficiently. The lack of an automatic system capable of distinguishing waste types in real-time hinders structured waste handling. This study aims to develop an automatic waste classification and sorting system based on a Convolutional Neural Network (CNN) using the YOLOv8 algorithm. The system is designed to recognize three main categories of waste organic, inorganic, and hazardous and toxic materials (B3) from images captured by a camera. The YOLOv8n model was trained using 1,000 images divided into ten object classes. The classification results are used to control a servo motor via an ESP32 microcontroller, directing waste to the appropriate disposal bin. The system is also equipped with a load cell sensor to measure weight and an ultrasonic sensor to detect waste volume. All classification data and sensor readings are displayed in real-time via an OLED screen and a MySQL-based web dashboard. Testing showed that the model achieved a precision of 1.00, with recall and F1-score ranging from 0.80 to 0.95. The load cell sensor demonstrated high accuracy with a maximum error of 0.20%, while the ultrasonic sensor recorded volume measurements with zero error. These results indicate that the system can classify and sort waste automatically, accurately, and responsively, with the potential to support more effective waste management in campus environments.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	CNN, Dashboard Web, Mikrokontroller, Klasifikasi Sampah, Sensor Berat, Sensor Ultrasonik, YOLOv8 .
Subjects:	Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques. Image analysis--Data processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	I Made Ricky Sudarnata
Date Deposited:	07 Aug 2025 08:44
Last Modified:	07 Aug 2025 08:44
URI:	http://repository.its.ac.id/id/eprint/127948

Actions (login required)

View Item