PUTRA, ANDIKA GUSTI RESTU (2026) DETEKSI KUALITAS BUAH APEL BERDASARKAN TEKSTUR DAN WARNA MENGGUNAKAN VISION TRANSFORMER (ViT) DAN GRADIENT BOOSTING. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
Cover.pdf Download (413kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
Bab 1.pdf Restricted to Registered users only Download (52kB) |
|
|
Text (BAB II)
Bab 2.pdf Restricted to Registered users only Download (235kB) |
|
|
Text (BAB III)
Bab 3.pdf Restricted to Registered users only Download (242kB) |
|
|
Text (BAB IV)
Bab 4.pdf Restricted to Registered users only Download (188kB) |
|
|
Text (BAB V)
Bab 5.pdf Restricted to Registered users only Download (46kB) |
|
|
Text (DAFTAR PUSTAKA)
Daftar Pustaka.pdf Restricted to Registered users only Download (96kB) |
|
|
Text (LAMPIRAN)
Lampiran.pdf Restricted to Registered users only Download (433kB) |
Abstract
Apple quality is one of the key factors influencing market value and consumer satisfaction; therefore, the sorting process must be carried out accurately and consistently. However, the quality assessment process—which is still performed manually—often relies on individual experience and visual observation, potentially leading to inconsistent results and reduced efficiency when sorting large quantities of fruit. This study aims to develop a classification model capable of detecting apple quality into two categories—fresh and rotten—based on texture and color characteristics in digital images. The approach combines texture features extracted using the Gray-Level Co-occurrence Matrix (GLCM), RGB-based color features, and high-level visual features obtained from a pretrained Vision Transformer (ViT) model without fine-tuning. All the resulting features were then combined and classified using the Gradient Boosting Classifier algorithm. The model’s performance was evaluated using a confusion matrix with metrics such as accuracy, precision, recall, and F1-score to assess the model’s ability to classify apple quality. Additionally, the developed model was implemented into the web�based application “Apple Quality AI” so it can be used to automatically detect apple quality. This research is expected to provide a faster, more consistent, and more objective solution compared to manual sorting processes and to serve as an alternative application of computer vision and machine learning technologies in supporting the quality control of agricultural products. Keywords: Apple Quality, Vision Transformer, Gradient Boosting, GLCM, RGB, Computer Vision. Kualitas buah apel menjadi salah satu faktor penting yang memengaruhi nilai jual dan tingkat kepuasan konsumen, sehingga proses penyortiran perlu dilakukan secara tepat dan konsisten. Namun, proses penilaian kualitas yang masih dilakukan secara manual sering kali bergantung pada pengalaman serta pengamatan visual masing-masing individu, sehingga berpotensi menimbulkan perbedaan hasil dan menurunkan efisiensi ketika jumlah buah yang disortir cukup banyak. Penelitian ini bertujuan untuk mengembangkan model klasifikasi yang mampu mendeteksi kualitas buah apel ke dalam dua kategori, yaitu fresh (segar) dan rotten (busuk), berdasarkan karakteristik tekstur dan warna pada citra digital. Pendekatan yang digunakan mengombinasikan fitur tekstur yang diekstraksi menggunakan Gray�Level Co-occurrence Matrix (GLCM), fitur warna berbasis statistik RGB, serta fitur visual tingkat tinggi yang diperoleh dari model Vision Transformer (ViT) pretrained tanpa proses fine-tuning. Seluruh fitur yang dihasilkan kemudian digabungkan dan diklasifikasikan menggunakan algoritma Gradient Boosting Classifier. Performa model dievaluasi menggunakan confusion matrix dengan metrik accuracy, precision, recall, dan F1-score untuk mengetahui kemampuan model dalam mengklasifikasikan kualitas buah apel. Selain itu, model yang telah dikembangkan diimplementasikan ke dalam aplikasi berbasis web Apple Quality AI sehingga dapat digunakan untuk melakukan deteksi kualitas buah apel secara otomatis. Penelitian ini diharapkan dapat memberikan solusi yang lebih cepat, konsisten, dan objektif dibandingkan proses sortasi manual serta menjadi salah satu alternatif penerapan teknologi computer vision dan machine learning dalam mendukung pengendalian mutu hasil pertanian. Kata kunci: Kualitas Buah Apel, Vision Transformer, Gradient Boosting, GLCM, RGB, Computer Vision.
Actions (login required)
 |
View Item |
