Rancang Bangun Prototipe Konveyor Seleksi Tingkat Kematangan Buah Tomat Secara Otomatis Menggunakan TinyML Berbasis ESP32-CAM

Tauristino, Mochammad Gery (2026) Rancang Bangun Prototipe Konveyor Seleksi Tingkat Kematangan Buah Tomat Secara Otomatis Menggunakan TinyML Berbasis ESP32-CAM. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5024221048-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (6MB) | Request a copy

Abstract

Sortasi buah tomat secara manual, khususnya untuk menentukan tingkat kematangan, merupakan proses yang lambat, padat karya, dan sangat subjektif, yang menyebabkan inkonsistensi kualitas. Penelitian ini bertujuan untuk merancang dan membangun sebuah prototipe konveyor sortir otomatis yang berfokus pada klasifikasi buah tomat. Tantangan utama dalam penelitian ini adalah menjalankan model computer vision yang kompleks pada perangkat keras berdaya rendah dan berbiaya rendah. Untuk mengatasi hal ini, penelitian menerapkan metode TinyML (Tiny Machine Learning) guna mengoptimalkan dan mengimplementasikan model Convolutional Neural Network (CNN) agar dapat berjalan secara efisien langsung pada \emph{embedded system} ESP32-CAM. Sistem ini terdiri dari konveyor yang digerakkan motor DC, ESP32-CAM yang bertugas melakukan akuisisi citra sekaligus inferensi model on-device, dan motor servo sebagai aktuator pemilah. Hasil klasifikasi akan digunakan untuk menggerakkan motor servo dan memisahkan tomat ke wadah yang telah ditentukan. Hasil yang diharapkan adalah sebuah prototipe fungsional yang mendemonstrasikan kelayakan sistem sortasi cerdas end-to-end yang akurat, efisien, dan terjangkau.
==============================================================================================================================================
Manual sorting of tomato fruits, particularly for determining ripeness levels, is a slow, labor-intensive, and highly subjective process, leading to inconsistent quality. This research aims to design and develop an automated sorting conveyor prototype focused on tomato fruit classification. The main challenge in this study is running complex computer vision models on low-power and low-cost hardware. To address this, the research implements TinyML (Tiny Ma chine Learning) methods to optimize and deploy a Convolutional Neural Network (CNN) model efficiently on the ESP32-CAM embedded system. The system consists of a conveyor driven by a DC motor, an ESP32-CAM responsible for image acquisition and on-device model inference, and a servo motor as the sorting actuator. The classification results (e.g., Ripe, Semi-Ripe, Unripe, Rotten, and Not a Tomato) will be used to actuate the servo motor and separate the tomatoes into designated containers. The expected outcome is a functional prototype demon strating the feasibility of an accurate, efficient, and affordable end-to-end intelligent sorting system.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Sortir Otomatis, TinyML, ESP32-CAM, Convolutional Neural Network (CNN), Klasifikasih Buah Tomat, Automated Sorting, Tomatot Fruit Classification
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > Q Science (General) > Q337.5 Pattern recognition systems Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.6 Computer programming. Q Science > QA Mathematics > QA76.758 Software engineering Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) Q Science > QA Mathematics > QA76 Computer software > QA76.8 Microprocessor T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User:	Mochammad Gery Tauristino
Date Deposited:	08 Jul 2026 05:26
Last Modified:	08 Jul 2026 06:15
URI:	http://repository.its.ac.id/id/eprint/134402

Actions (login required)

View Item