CD Skripsi

Klasifikasi Tingkat Kematangan Buah Pisang Cavendish Menggunakan Convolutional Neural Network

---

The Cavendish banana is an important agricultural commodity in the farming industry. The maturity level of Cavendish bananas is a critical factor influencing the fruit's durability and quality. A Convolutional Neural Network (CNN) method was used in this study to facilitate the classification of Cavendish banana maturity levels. The model utilized 800 training data points, 200 validation data points, under two conditions: original and modified layers, with the modified condition including four additional dense layers. Six CNN models were developed using three base model architectures: MobileNet, DenseNet201, and InceptionV3. The batch size used was 32, with an image size of 224 x 224, and a size of 299 x 299 specifically for InceptionV3. The model’s optimizer was configured with Adam, using a learning rate of 0.001 and Early Stopping as a callback. Model training was conducted with a maximum of 50 epochs and 25 steps per epoch. The coding was implemented in Python using the TensorFlow library. Based on test results, the DenseNet201 model achieved better accuracy, and precision than other models in the original, with highest accuracy of 0.9444, and the average precision in the original condition was higher at 0.9466, compared to the modified condition. Therefore, the CNN-based DenseNet201 model can be used as an effective tool for classifying the maturity level of Cavendish bananas.

Keywords: Classification, Maturity Level, Convolutional Neural Network, Cavendish Banana, CNN Model

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2003113202

Penerbit

Pekanbaru : Universitas Riau FMIPA Sistem Informasi., 2024

Deskripsi Fisik

-

Bahasa

ISBN/ISSN

-

Klasifikasi

2003113202

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