Baihaqi, Rafli Akhmad (2024) Penerapan Convolutional Neural Network Dalam Rekonstruksi Gambar Ground Penetrating Radar Dua Dimensi Yang Rusak. Other thesis, Institut Teknologi Sepuluh Nopember.
![]() |
Text
5024201054-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (10MB) | Request a copy |
Abstract
Ground Penetrating Radar (GPR) adalah alat non-destruktif untuk menjelajahi objek bawah permukaan menggunakan gelombang elektro magnetik. Gelombang terdistorsi oleh perbedaan sifat elektromagnetik, seperti permitivitas dan konduktivitas, memungkinkan identifikasi karakteristik benda yang terkubur. Hasil sinyal yang diperoleh dari Ground Penetrating Radar (GPR) seringkali mengalami tantangan seperti hilangnya informasi atau pergeseran dalam berbagai situasi atau kondisi tertentu. Hal ini dapat disebabkan oleh beberapa faktor, termasuk jenis material yang ada di bawah permukaan. Material dengan konduktivitas listrik tinggi, seperti tanah yang sangat lembab atau air, cenderung menyerap lebih banyak energi gelombang GPR, yang pada akhirnya dapat mengurangi kedalaman maksimum yang dapat dijangkau oleh sinyal GPR dan membuat hasil sinyal kurang terlihat dengan jelas. Penelitian ini memperkenalkan solusi menggunakan deep learning untuk mengisi data yang hilang pada gambar. Arsitektur yang akan digunakan adalah Convolutional Autoencoder (CAE) karena dapat mengenali fitur dari suatu gambar dan melakukan prediksi berdasarkan fitur yang telah dipelajari
==================================================================================================================================
Ground Penetrating Radar (GPR) is a non-destructive tool for exploring subsurface objects using electromagnetic waves. The waves are distorted by differences in electromagnetic properties, such as permittivity and conductivity, allowing identification of the characteristics of buried objects. The signal results obtained from Ground Penetrating Radar (GPR) often experience challenges such as loss of information or shifts in various situations or certain conditions. This can be caused by several factors, including the type of material below the surface. Materials with high electrical conductivity, such as very moist soil or water, tend to absorb more GPR wave energy, which in turn can reduce the maximum depth a GPR signal can reach and make the resulting signal less clearly visible. This research introduces a solution using deep learning to fill in missing data in images. The architecture that will be used is Convolutional Autoencoder (CAE) because it can recognize features from an image and make predictions based on the features that have been learned.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | GPR, Autoencoder, Deep Learning, Missing Data GPR,Autoencoder,Deep Learning, Data Hilang |
Subjects: | T Technology > T Technology (General) > T11 Technical writing. Scientific Writing T Technology > T Technology (General) > T385 Visualization--Technique T Technology > T Technology (General) > T57.5 Data Processing T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis |
Depositing User: | Rafli Akhmad Baihaqi |
Date Deposited: | 25 Jul 2024 05:56 |
Last Modified: | 25 Jul 2024 05:56 |
URI: | http://repository.its.ac.id/id/eprint/108736 |
Actions (login required)
![]() |
View Item |