Implementasi algoritma Convolutional Neural Network (CNN) dengan arsitektur mobilenetv2 pada klasifikasi jenis tanaman obat berdasarkan citra daun

Alfian, Moch Rizky (2025) Implementasi algoritma Convolutional Neural Network (CNN) dengan arsitektur mobilenetv2 pada klasifikasi jenis tanaman obat berdasarkan citra daun. Sarjana thesis, UIN Sunan Gunung Djati Bandung.

Preview	Text (COVER) 1_cover.pdf Download (204kB) | Preview
Preview	Text (ABSTRAK) 2_abstrak.pdf Download (245kB) | Preview
Preview	Text surat pernyataan karya sendiri Moch Rizki Alfian.pdf Download (165kB) | Preview
Preview	Text (DAFTAR ISI) 4_daftarisi.pdf Download (237kB) | Preview
Preview	Text (BAB I) 5_bab1.pdf Download (262kB) | Preview
	Text (BAB II) 6_bab2.pdf Restricted to Registered users only Download (616kB) | Request a copy
	Text (BAB III) 7_bab3.pdf Restricted to Registered users only Download (594kB) | Request a copy
	Text (BAB IV) 8_bab4.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text (BAB V) 9_bab5.pdf Restricted to Registered users only Download (199kB) | Request a copy
	Text (DAFTAR PUSTAKA) 10_daftarpustaka.pdf Restricted to Registered users only Download (204kB) | Request a copy
	Text (LAMPIRAN) 11_lampiran.pdf Restricted to Repository staff only Download (226kB) | Request a copy

Abstract

Medicinal plants play an essential role in traditional medicine as they contain compounds with various pharmacological benefits, such as antimicrobial, anticancer, and anti-inflammatory properties. Indonesia, as a country with high biodiversity, has numerous medicinal plant species used for disease prevention and treatment. However, manually identifying these plants remains a challenge as it requires time and extensive botanical expertise. To address this issue, this study applies the CRISP-DM (Cross-Industry Standard Process for Data Mining) methodology as a framework for data processing, consisting of six main stages: business understanding, data understanding, data preparation,odeling, evaluation, and deployment.In the modeling stage, the MobileNetV2 architecture, a lightweight and efficient CNN model, is utilized to classify medicinal plants based on leaf images. The model is tested with various data split ratios (60:40, 70:30, 80:20) and different epoch values (20, 50, 80, 100). The results show that MobileNetV2 achieves the highest accuracy at a 60:40 ratio, with an accuracy of 99.90%, precision of 99.90%, recall of 99.90%, and F1-score of 99.90%, demonstrating its effectiveness in classifying medicinal plant species based on leaf images.

Actions (login required)

View Item