PENERAPAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN) UNTUK IDENTIFIKASI KEKURANGAN UNSUR HARA PADA TANAMAN TIMUN APEL BERBASIS PENGOLAHAN CITRA

Agus Wicaksono, Mardika and Qur’ania, Arie and Herdianto Situmorang, Boldson (2019) PENERAPAN METODE CONVOLUTIONAL NEURAL NETWORK (CNN) UNTUK IDENTIFIKASI KEKURANGAN UNSUR HARA PADA TANAMAN TIMUN APEL BERBASIS PENGOLAHAN CITRA. Skripsi thesis, Universitas Pakuan.

	Text COVER.pdf Download (36kB)
	Text PENGESAHAN.pdf Download (165kB)
	Text PENGESAHAN.pdf Download (165kB)
	Text DAFTAR PUSTAKA.pdf Download (42kB)

Abstract

Abstract Nutrients are important components that are needed by plants for growth and development. One part of the plant that can be used as an indicator of nutrients is the leaves. In Indonesia, various types of plants are spread throughout the region. Not a few of these plants are the result of marriage between plants with one another, for example: timun apel plants. Timun apel is a type of fruit that results from a crossing between Timun suri (Cucumis sativus) and Melon (Cucumis melo). This study aims to apply the Convolutional Neural Network (CNN) method to identify nutrient deficiencies in timun apel plants based on image processing. The Convolutional Neural Network (CNN) method seeks to mimic the way of recognizing the same image patterns of connections of neurons or nerve cells in the human brain so as to be able to process the same information. The data of this study amounted to 100 images of timun apel leaves which were divided using the k-fold cross validation method with a value of k = 4 so as to produce 75 training data and 25 testing data. The output of identification in this study were 5 criteria: nitrogen deficiency, phosphorus deficiency, potassium deficiency, nitrogen deficiency, phosphorus and potassium and normal. The test is carried out 5 times which results in an average accuracy of 74.2%. The highest accuracy of 77.0% resulted from the third test. While the lowest accuracy of 72.6% resulted from the first test and fifth test. The results of this study indicate that the CNN architecture can identify 5 nutrient deficiency criteria automatically. Keywords : Nutrients, Timun Apel, Image Processing, Convolutional Neural Network (CNN)

Item Type:	Thesis (Skripsi)
Subjects:	Fakultas Ilmu Pengetahuan Alam dan Matematika > Ilmu Komputer
Divisions:	Fakultas Matematika dan Ilmu Pengetahuan Alam > Ilmu Komputer
Depositing User:	PERPUSTAKAAN FAKULTAS MATEMATIKA DAN ILMU PENGETAHUAN ALAM UNPAK
Date Deposited:	01 Sep 2022 02:30
Last Modified:	22 Sep 2022 07:27
URI:	http://eprints.unpak.ac.id/id/eprint/3925

Actions (login required)

View Item