Klasifikasi spesies tumbuhan invasif menggunakan algoritma convolutional neural network (CNN)

Apriyanti, Dhea Listia (2025) Klasifikasi spesies tumbuhan invasif menggunakan algoritma convolutional neural network (CNN). Sarjana thesis, UIN Sunan Gunung Djati Bandung.

Preview	Text 1_cover.pdf Download (189kB) | Preview
Preview	Text 2_abstrak.pdf Download (205kB) | Preview
Preview	Text 3_skbebasplagiarism.pdf Download (286kB) | Preview
Preview	Text 4_daftarisi.pdf Download (264kB) | Preview
Preview	Text 5_bab1.pdf Download (384kB) | Preview
	Text 6_bab2.pdf Restricted to Registered users only Download (451kB) | Request a copy
	Text 7_bab3.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text 8_bab4.pdf Restricted to Registered users only Download (2MB) | Request a copy
	Text 9_bab5.pdf Restricted to Registered users only Download (222kB) | Request a copy
	Text 10_daftarpustaka.pdf Restricted to Registered users only Download (220kB) | Request a copy

Abstract

INDONESIA: Tumbuhan invasif merupakan spesies yang dapat mengganggu ekosistem dan mengancam keanekaragaman hayati di lingkungan sekitarnya. Oleh karena itu, identifikasi dini terhadap spesies tumbuhan invasif sangat penting untuk dilakukan. Penelitian ini bertujuan untuk mengembangkan model klasifikasi spesies tumbuhan invasif menggunakan algoritma Convolutional Neural Network (CNN) dengan arsitektur EfficientNet-B1. Dataset terdiri dari 3000 gambar yang dibagi ke dalam tiga subset, yaitu data latih, validasi, dan uji, dengan beberapa variasi rasio pembagian yaitu, 80:10:10 dan 70:15:15. Proses pelatihan dilakukan pada ukuran gambar 240x240 piksel dan jumlah epoch yang divariasikan antara 20, 40, dan 60. Hasil evaluasi terbaik diperoleh pada rasio 80:10:10 dengan epoch 60, yang menghasilkan nilai akurasi sebesar 97.33%, precision 97.57%, recall 97.33%, dan f1-score 97.32%. Hasil dari penelitian ini menunjukkan bahwa model mampu mengklasifikasikan spesies tumbuhan invasif secara efektif dan akurat, serta dapat menjadi solusi awal dalam mendukung upaya pelestarian lingkungan. ENGLISH: Invasive plants are species that can disrupt ecosystems and threaten biodiversity in their surrounding environment. Therefore, early identification of invasive plant species is very important. This study aims to develop a classification model for invasive plant species using the Convolutional Neural Network (CNN) algorithm with the EfficientNet-B1 architecture. The dataset consists of 3,000 images divided into three subsets: training, validation, and testing, with several variations in the division ratio, namely 80:10:10 and 70:15:15. The training process was conducted on images of 240x240 pixels, with the number of epochs varied between 20, 40, and 60. The best evaluation results were obtained at a ratio of 80:10:10 with 60 epochs, yielding an accuracy of 97.33%, precision of 97.57%, recall of 97.33%, and an F1-score of 97.32%. The results of this study indicate that the model is capable of effectively and accurately classifying invasive plant species and can serve as an initial solution to support environmental conservation efforts.

Actions (login required)

View Item