Islam, Ikhbal Ihwanul (2026) RANCANG BANGUN APLIKASI PENDETEKSI SPESIES POHON AGLAONEMA MENGGUNAKAN ALGORITMA CONVOLUTIONAL NEURAL NETWORK (CNN) (STUDI KASUS : RUMAH BUNGA CILIMUS - CARACAS). S1 / D3 thesis, Universitas Kuningan.

[thumbnail of ABSTRAK] Text (ABSTRAK)
ABSTRAK.pdf

Download (3MB)
[thumbnail of BAB I] Text (BAB I)
BAB I.pdf

Download (686kB)
[thumbnail of BAB II] Text (BAB II)
BAB II.pdf
Restricted to Repository staff only

Download (954kB) | Request a copy
[thumbnail of BAB III] Text (BAB III)
BAB III.pdf
Restricted to Repository staff only

Download (1MB) | Request a copy
[thumbnail of BAB IV] Text (BAB IV)
BAB IV.pdf
Restricted to Repository staff only

Download (839kB) | Request a copy
[thumbnail of BAB V] Text (BAB V)
BAB V.pdf

Download (336kB)
[thumbnail of DAFTAR PUSTAKA] Text (DAFTAR PUSTAKA)
DAFTAR PUSTAKA.pdf

Download (143kB)
[thumbnail of LAMPIRAN] Text (LAMPIRAN)
LAMPIRAN.pdf
Restricted to Repository staff only

Download (2MB) | Request a copy
Official URL: https://rama.uniku.ac.id

Abstract

Perkembangan teknologi kecerdasan buatan, khususnya di bidang computer vision, memberikan peluang besar dalam pemanfaatan model pembelajaran mesin untuk identifikasi objek secara otomatis. Penelitian ini bertujuan untuk membangun aplikasi pendeteksi spesies Aglaonema berbasis Android yang dapat membantu pengguna, khususnya pengunjung Rumah Bunga Kuningan, dalam mengenali jenis-jenis Aglaonema secara cepat dan akurat. Proses pengembangan sistem meliputi pengumpulan dataset citra Aglaonema, preprocessing, augmentasi data, perancangan model menggunakan arsitektur MobileNetV2, serta pelatihan model dengan teknik transfer learning. Model yang telah dilatih kemudian diintegrasikan ke dalam aplikasi Android menggunakan TensorFlow Lite agar dapat dijalankan secara efisien pada perangkat seluler. Pengujian akurasi menunjukkan bahwa model mampu mengidentifikasi spesies dengan performa yang stabil pada data pelatihan maupun validasi. Evaluasi aplikasi dilakukan menggunakan metode User Acceptance Test (UAT) dan menunjukkan bahwa sistem dapat diterima pengguna karena mudah dioperasikan dan memberikan hasil deteksi yang akurat. Secara keseluruhan, aplikasi ini berhasil menjadi solusi pendukung edukasi dan identifikasi tanaman hias Aglaonema serta berpotensi dikembangkan lebih lanjut dengan peningkatan jumlah dataset dan fitur pendukung lainnya.

The rapid advancement of artificial intelligence, particularly in computer vision,
has enabled the development of automated object recognition systems that can
operate efficiently on mobile devices. This research focuses on building an
Android-based application capable of detecting Aglaonema plant species to assist
users, especially visitors of Rumah Bunga Kuningan, in identifying various
Aglaonema types quickly and accurately. The development process includes dataset
collection, preprocessing, data augmentation, and model training using the
MobileNetV2 architecture with a transfer learning approach. The trained model is
then converted into TensorFlow Lite format for lightweight execution on Android
smartphones. Evaluation results show that the model achieves stable and high
accuracy in both training and validation datasets. Furthermore, user acceptance
testing (UAT) indicates that the application is easy to use, functions properly, and
provides accurate detection results according to user expectations. Overall, this
application successfully serves as a practical tool for plant identification and
education, with potential future improvements through expanding the dataset and
adding enhanced features.

Item Type: Thesis (S1 / D3)
Uncontrolled Keywords: Aglaonema, Deteksi Citra, MobileNetV2, TensorFlow, Android.
Subjects: T Technology > T Technology (General)
Divisions: Fakultas Ilmu Komputer > S1 Teknik Informatika
Depositing User: S.Kom Ikhbal Ihwanul Islam
Date Deposited: 06 Jul 2026 07:31
Last Modified: 06 Jul 2026 07:31
URI: https://rama.uniku.ac.id/id/eprint/5437

Actions (login required)

View Item
View Item