NURFADLILLAH, MAHALBI (2026) KLASIFIKASI TINGKAT KEMATANGAN PISANG PADA CITRA DIGITAL MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK (CNN). S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
Cover.pdf Download (545kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
Bab 1.pdf Restricted to Registered users only Download (46kB) |
|
|
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 (186kB) |
|
|
Text (BAB IV)
Bab 4.pdf Restricted to Registered users only Download (528kB) |
|
|
Text (BAB V)
Bab 5.pdf Restricted to Registered users only Download (26kB) |
|
|
Text (DAFTAR PUSTAKA)
Daftar pustaka.pdf Restricted to Registered users only Download (150kB) |
|
|
Text (LAMPIRAN)
Lampiran.pdf Restricted to Registered users only Download (664kB) |
Abstract
Bananas as a major horticultural commodity in Indonesia, contribute significant economic value and play a strategic role in maintaining food stability at the national level. One important aspect in maintaining the quality and competitiveness of bananas is determining the correct level of ripeness, as it significantly affects taste, nutritional content, and market value. Traditional methods for determining banana ripeness still rely heavily on direct eye observation, which is subjective and potentially inaccurate, especially on a large scale. In line with current innovation advances, the implementation of digital image processing technology and the integration of artificial intelligence, particularly deep learning, are providing more accurate, objective, and consistent solutions. Research to classify banana ripeness levels with three ripeness classes using a Convolutional Neural Network (CNN) architecture with geometric augmentation achieved a test accuracy of 93.28% and a training accuracy of 92.29%. This architectural model is based on its effectiveness in data processing and its capable categorization capabilities for moderate-scale visual data sets. This study is projected to contribute new ideas for the advancement of an automatic fruit identification system that will not only help farmers and business actors but also strengthen the bargaining power of domestic banana products in the international trade arena. Kata kunci: Maturity Classification, Digital Image, Convolutional Neural Network (CNN) Pisang sebagai komoditas hortikultura utama di tanah air, pisang menyumbang nilai ekonomis yang signifikan serta berperan strategis dalam menjaga stabilitas pangan di tingkat nasional. Salah satu aspek penting dalam menjaga kualitas dan daya saing pisang adalah penentuan tingkat kematangan yang tepat, karena sangat mempengaruhi cita rasa, kandungan nutrisi, serta nilai jual di pasar. Metode tradisional penentuan kemasakan buah pisang masih banyak lazimnya masih bertumpu pada observasi mata secara langsung, sehingga bersifat subjektif dan berpotensi menimbulkan ketidakakuratan, terutama dalam skala besar. Sejalan dengan kemajuan inovasi saat ini, implementasi teknologi pemrosesan gambar digital serta integrasi kecerdasan artifisial, khususnya deep learning, menjadi solusi yang lebih akurat, objektif, dan konsisten. Penelitian untuk mengklasifikasikan tingkat kematangan pisang dengan tiga kelas kematangan menggunakan arsitektur Convolutional Neural Network (CNN) dengan augmentasi geometric akurasi uji mencapai 93.28% dan latih 92.29%. Model arsitektur ini didasarkan pada efektivitasnya dalam pemrosesan data serta kemampuan kategorisasi yang mumpuni untuk kumpulan data visual berskala moderat. Studi ini diproyeksikan mampu menyumbangkan pemikiran baru bagi kemajuan sistem identifikasi buah otomatis yang tidak hanya membantu petani dan pelaku usaha melainkan juga memperkokoh nilai tawar produk pisang domestik di kancah perdagangan internasional. Kata kunci: Klasifikasi Kematangan, Citra Digital, Convolutional Neural Network (CNN)
Actions (login required)
 |
View Item |
