Responsive Banner

Identifikasi lahan kosong Kota Batu berbasis Citra Google Earth menggunakan Algoritma Convolutional Neural Network (CNN)

Alam, Islam Nur (2020) Identifikasi lahan kosong Kota Batu berbasis Citra Google Earth menggunakan Algoritma Convolutional Neural Network (CNN). Undergraduate thesis, Universitas Islam Negeri Maulana Malik Ibrahim.

This is the latest version of this item.

[img]
Preview
Text (Fulltext)
16650096.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB) | Preview

Abstract

ABSTRACT

Identification of vacant land based on Google Earth imagery is axresearch materialxthat is currentlyxbeingdeveloped. The details of the characteristics that must be captured make researchers competing to find the mostxsuitable method forxidentification. The Convolutional Neural Network (CNN) algorithmxisxone ofxthexalgorithmsxcurrentlyxmost favoredxin thexfield ofxclassification and identification of objects at this time. The data used in this study there is data in the form of image cropping results from Google Earth imagery. Before the data is trained, the data will be preprocessed, including crop, using an RGB image, and then segmented. After that the algorithm will be trained using 1300 images divided into 2 classes. With an image size of 32x32 pixels, then trained with 1000 epochs and a learning rate of 0.001. After that, the model that has been trainedis tested on 300 new images. From these studies the results obtained in the form of training accuracy of 99.6% and testing accuracy of 86.34%

ABSTRAK

Identifikasi lahan kosong Kota Batu berbasis citra google earth adalah materi penelitian yang saat ini terus dikembangkann. Detailnya ciri yang mesti ditangkap membuat para peneliti berlomba-lomba menemukan metode yang paling cocok untuk melakukan identifikasi. Algoritma Convolutional Neural Network (CNN) menjadi salah satu algoritma yang saat ini paling diunggulkan dalam bidang klasifikasi dan identifikasi objek saat ini. Data yang digunakan pada penelitianxini ada data berupa citra hasil cropping dari citra google earth. Sebelum data dilatih, data akan dilakukan preprocessing antara lain crop, menggunakan citra RGB, kemudian diaugmentasi. Setelah itu algoritma akan dilatih menggunakan citra sebanyak 1300 citra yang dibagi menjadi 2 kelas. Dengan ukuran citra sebesar 32x32 pixel, kemudian dilatih dengan 1000 epoch dan learning rate 0.001. Setalah itu, model yang sudah dilatih dilakukan pengujian pada 300 citra baru. Dari penelitian tersebut didapatkan hasil berupa akurasi training sebesar 99,6% dan akurasi testinggsebesar 86,34%.

Item Type: Thesis (Undergraduate)
Supervisor: Santoso, Irwan Budi and Fatchurrochman, Fatchurrochman
Contributors:
ContributionNameEmail
UNSPECIFIEDSantoso, Irwan BudiUNSPECIFIED
UNSPECIFIEDFatchurrochman, FatchurrochmanUNSPECIFIED
Keywords: neural nets; convolution; preprocessing; Google Earth image; training; testing; data augmentation; neural nets; convolution; preprocessing; Google Earth image; training; testing; data augmentation
Subjects: 08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080199 Artificial Intelligence and Image Processing not elsewhere classified
Departement: Fakultas Sains dan Teknologi > Jurusan Teknik Informatika
Depositing User: Islam Nur Alam
Date Deposited: 22 Jul 2020 09:36
Last Modified: 02 May 2023 09:55
URI: http://etheses.uin-malang.ac.id/id/eprint/19840

Downloads

Downloads per month over past year

Available Versions of this Item

  • Identifikasi lahan kosong Kota Batu berbasis Citra Google Earth menggunakan Algoritma Convolutional Neural Network (CNN). (deposited 22 Jul 2020 09:36) [Currently Displayed]

Actions (login required)

View Item View Item