Junaedi, Jujun (2025) RANCANG BANGUN APLIKASI KLASIFIKASI JENIS TANAMAN HERBAL MENGGUNAKAN ALGORITMA CONVOLUTIONAL NEURAL NETWORK (CNN) (Studi Kasus: Kios Bunga Rabiku Florist). S1 / D3 thesis, Universitas Kuningan.

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

Download (698kB)
[thumbnail of BAB 1] Text (BAB 1)
BAB I.pdf

Download (607kB)
[thumbnail of BAB II] Text (BAB II)
BAB II.pdf
Restricted to Repository staff only

Download (1MB) | 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 (2MB) | Request a copy
[thumbnail of BAB V] Text (BAB V)
BAB V.pdf

Download (187kB)
[thumbnail of DAFTAR PUSTAKA] Text (DAFTAR PUSTAKA)
DAFTAR PUSTAKA.pdf

Download (166kB)
[thumbnail of LAMPIRAN] Text (LAMPIRAN)
LAMPIRAN.pdf
Restricted to Repository staff only

Download (4MB) | Request a copy

Abstract

Tanaman herbal memiliki peran penting dalam kehidupan manusia karena kandungan senyawa aktifnya yang bermanfaat bagi kesehatan. Banyaknya jenis tanaman herbal yang memiliki karakteristik fisik yang serupa sering kali menyulitkan pembeli dan masyarakat awam dalam mengidentifikasinya. Penelitian ini bertujuan untuk merancang dan membangun aplikasi klasifikasi jenis tanaman herbal berbasis Android menggunakan algoritma Convolutional Neural Network (CNN) dengan pendekatan transfer learning melalui arsitektur EfficientNetV2-S. Sistem dikembangkan untuk mengklasifikasikan 10 jenis tanaman herbal dengan total 10.000 gambar, dimana masing-masing kelas terdiri dari 1.000 gambar yang telah melalui proses preprocessing dan augmentasi. Pelatihan model dilakukan menggunakan Jupyter Notebook dengan akurasi klasifikasi sebesar 98%. Evaluasi menunjukkan precision, recall, dan f1-score yang tinggi, seperti Jarak Tintir dan Temulawak dengan f1-score 1.00, serta rata-rata makro dan weighted average sebesar 0.98. Aplikasi memungkinkan pengguna mengklasifkasi tanaman melalui kamera, galeri, maupun secara real-time, dan menampilkan nama serta manfaatnya secara otomatis. Hasil penelitian menunjukkan bahwa CNN mampu memberikan klasifikasi yang akurat, serta aplikasi yang dibangun dapat membantu pengguna, khususnya di Rabiku Florist, mengenali tanaman herbal secara cepat, tepat, dan praktis. Penelitian ini diharapkan menjadi kontribusi nyata dalam digitalisasi pengetahuan tanaman herbal di Indonesia.
Herbal plants have an important role in human life because of their active compounds that are beneficial for health. The many types of herbal plants that have similar physical characteristics often make it difficult for buyers and ordinary people to identify them. This research aims to design and build an Android-based herbal plant classification application using the Convolutional Neural Network (CNN) algorithm with a transfer learning approach through the EfficientNetV2-S architecture. The system is developed to classify 10 types of herbal plants with a total of 10,000 images, where each class consists of 1,000 images that have gone through preprocessing and augmentation processes. Model training was conducted using Jupyter Notebook with a classification accuracy of 98%. Evaluation showed high precision, recall, and f1-score, such as Jarak Tintir and Temulawak with f1-score 1.00, and macro and weighted average of 0.98. The application allows users to classify plants through the camera, gallery, or in real-time, and displays their names and benefits automatically. The results showed that CNN is able to provide accurate classification, and the application built can help users, especially at Rabiku Florist, recognize herbal plants quickly, precisely, and practically. This research is expected to be a real contribution to the digitization of herbal plant knowledge in Indonesia.

Item Type: Thesis (S1 / D3)
Uncontrolled Keywords: Android; CNN; EfficientNetV2-S; Klasifikasi Citra; Tanaman Herbal. Android; Convolutional Neural Network; EfficientNetV2-S; Herbal Plants; Image Classification;
Subjects: T Technology > T Technology (General)
Divisions: Fakultas Ilmu Komputer > S1 Teknik Informatika
Depositing User: S.Kom Jujun Junaedi
Date Deposited: 25 Aug 2025 03:22
Last Modified: 25 Aug 2025 03:22
URI: https://rama.uniku.ac.id/id/eprint/3204

Actions (login required)

View Item
View Item