CD Skripsi

Identifikasi Tanaman Herbal Pembebas Nyeri Menggunakan Metode Convolution Neural Network Berbasis Arsitektur Mobilenetv2

---

Indonesia has abundant biodiversity, including more than 2,000 species of herbal plants that have the potential to be used as natural medicines. However, people still tend to prefer chemical drugs due to a lack of understanding of the benefits of herbal plants, including their ability to relieve pain. Additionally, many herbs are found in hard-to-reach areas, adding to the challenges in the identification process. This study aims to develop a model using the Convolutional Neural Network (CNN) method with MobileNetV2 architecture to identify pain-relieving herbal plants based on leaf images. The research process began with the collection of leaf image data from five types of pain-relieving herbal plants, namely star fruit, guava, aloe vera, pandanus, and betel. The data is processed through pre-processing stages such as resizing, normalizing and augmentation to improve the quality and variety of data. The CNN MobileNetV2 model is trained to use this data, and evaluated using metrics such as accuracy, precision, recall, and f1-score. The results showed that the modified model managed to achieve an accuracy of up to 99%, higher than the base model which had an accuracy of 98%. These findings reinforce that the MobileNetV2 architecture is effective in automatically identifying pain-relieving herbal plants, potentially improving public understanding and supporting natural herbal plant-based remedies.

Keywords: MobileNetV2, CNN, Medicinal Plants, Data Augmentation, Image Classification.

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2003125720

Penerbit

Pekanbaru : Universitas Riau FMIPA Sistem Informasi., 2024

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2003125720

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

SISTEM INFORMASI

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Mutia

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar