Deteksi Ketersediaan Lahan Parkir Mobil Menggunakan Metode Transfer Learning Dan Convolutional Neural Network

Abdurrasyid, Muhammad Faris (2022) Deteksi Ketersediaan Lahan Parkir Mobil Menggunakan Metode Transfer Learning Dan Convolutional Neural Network. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
05111540000147-Udergraduate_Thesis.docx - Accepted Version
Restricted to Repository staff only

Download (12MB) | Request a copy
[img] Text
05111540000147-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 April 2024.

Download (2MB) | Request a copy

Abstract

Rata-rata orang menggunakan waktunya yang cukup lama untuk mencari tempat parkir. Untuk meringankan dan mengurangi waktu mencari tempat parkir, dibuatkan sebuah Parking Guidance Information (PGI). PGI membutuhkan informasi yang akurat dan terbarui untuk menyediakan informasi yang dapat diandalkan oleh user dalam mencari sebuah lahan parkir. Sistem yang akan dibuat terdiri dari 4 tahapan yaitu, proses pengolahan data mentah, training CNN model, testing, dan pengukuran performa presisi, recall, dan akurasi dan bounding box sebagai keluaran informasi ketersediaan lahan parkir. Dalam uji coba performa terbaik didapatkan dengan menggunakan mini VGG 16. Mini VGG 16 merupakan arsitektur yang dibuat berdasarkan analisis dari uji coba menggunakan VGG 16 dan VGG 16 dengan transfer learning. Mini VGG 16 memiliki performa akurasi dan kecepatan yang lebih handal dibanding VGG 16 dan VGG 16 dengan transfer learning dengan akurasi 0.963. =============================================================================================== The average person spends 7.8 minutes looking for a parking space. To ease and reduce the time to find a parking space, a Parking Guidance Information (PGI) was made. PGI requires information that is accurate and up to date to provide information that can be relied on by users in finding a parking space. The system that will be made consists of 4 stages, namely, the processing of raw data, CNN training model, Testing, and performance measurement (Precision, Recall, and Accuracy) and Bounding Box as an output of parking space availability information. In testing the best performance is obtained by using a mini VGG 16. Mini VGG 16 is an architecture created based on the analysis of trials using VGG 16 and VGG 16 with transfer learning. Mini VGG 16 has more reliable performance and speed than VGG 16 and VGG 16 with transfer learning with accuracy 0.963.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Parking Guidance Information (PGI), Mini VGG 16, training CNN model, testing, pengukuran performa presisi, recall, akurasi bounding box
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning.
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User: Muhammad Faris Abdurrasyid
Date Deposited: 28 Mar 2022 02:24
Last Modified: 28 Mar 2022 02:24
URI: https://repository.its.ac.id/id/eprint/94877

Actions (login required)

View Item View Item