Rancang Bangun Sistem Pendeteksi Penyakit Pada Tanaman Stroberi Dengan Metode CNN

Dinata, Muhammad Imam (2021) Rancang Bangun Sistem Pendeteksi Penyakit Pada Tanaman Stroberi Dengan Metode CNN. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 07111950065003-Master_thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Stroberi merupakan tanaman buah favorit yang ditanam oleh para petani dataran tinggi karena cara menanamnya yang mudah. Namun tanaman buah ini sering terkena penyakit dimana petani hanya menduga penyakit apa yang menyerang tanamannya, sehingga pengobatannya masih sebatas menebak. Beberapa penelitian terhadap buah stroberi telah dilakukan yaitu pendeteksian buah stroberi dengan CNN, namun belum mampu mengatasi masalah utama penyakit pada buah stroberi. Untuk itu dibutuhkan sebuah sistem yang dapat mengetahui penyakit pada tanaman stroberi dengan memanfaatkan penerapan metode deep learning. Salah satu metode pada deep learning yang dapat digunakan yaitu metode convolutional neural network.
Metode Convolutional Neural Network (CNN) merupakan salah satu metode deep learning yang terkenal mampu mendeteksi gambar secara akurat. Dalam penelitian ini peneliti menggunakan metode CNN untuk mendeteksi jenis penyakit yang menginfeksi pada tanaman stroberi dan akan memberikan saran dan pencegahan. Sehingga diharapkan petani hanya perlu menggunakan smartphone androidnya untuk mengetahui jenis penyakit pada tanaman stroberi dan cara mengatasinya.
Dari hasil eksperimen ini, model CNN yang berhasil dibuat dengan tingkat akurasi sebesar 69%, dengan akurasi yang diperoleh tersebut dilakukan pendeteksian terhadap penyakit daun tanaman stroberi dengan nilai rata-rata sebesar 0.227 frame per second

Kata Kunci: Tanaman Stroberi, Penyakit Stroberi, CNN, Smartphone, Deteksi.

==========================================================

Strawberry is a favorite fruit crop planted by highland farmers because of how easy it is to grow. However, this fruit plant often gets diseases where farmers only guess what disease is infected on their plants, so the treatment is still limited to guessing. Some research on strawberry has been carried out, namely the detection of Strawberry with CNN, but has not been able to overcome the main problem of Strawberry disease. For that, need a system that can detect diseases in strawberry plants by utilizing the application of deep learning methods. One method in deep learning that can be used is the convolutional neural network method.
The Convolutional Neural Network (CNN) method is one of the well-known deep learning methods capable of accurately detecting images. In this research, the researcher uses the CNN method to detect the type of disease that is infected on strawberry plants and will provide suggestions. So that farmers are expected to only need to use their android smartphones to find out the types of diseases in strawberry plants and how to overcome them.
From the results of this experiment, the CNN model was successfully created with an accuracy rate of 69%, with the accuracy obtained, the detection of strawberry leaf disease was carried out with an average value of 0.227 frames per second.

Keywords: Strawberry Plants, Strawberry Disease, CNN, Smartphone, Detection.

Item Type:	Thesis (Masters)
Subjects:	S Agriculture > S Agriculture (General) T Technology > T Technology (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7882.P3 Pattern recognition systems
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	muhammad imam dinata
Date Deposited:	16 Aug 2021 03:34
Last Modified:	16 Aug 2021 03:34
URI:	http://repository.its.ac.id/id/eprint/87065

Actions (login required)

View Item