Article Information

Title: Deep Learning Classification of The Ground Penetrating Radar Data to Determine Underground Cavities Under the Road

Authors: Rohit Shrestha, Zhang Zhihou, Sulaiman Khan, Xiaoyan Zhao

Journal: Southern journal of engineering and technology (Print)

HEC Recognition History
No recognition records found.

Publisher:
Deprecated: htmlspecialchars(): Passing null to parameter #1 ($string) of type string is deprecated in /home/define/domains/articles.define.pk/public_html/article.php on line 456

Country:
Deprecated: htmlspecialchars(): Passing null to parameter #1 ($string) of type string is deprecated in /home/define/domains/articles.define.pk/public_html/article.php on line 460

Year: 2024

Volume: 1

Issue: 1

Language: en

Keywords: AlexNetCavitiesCNNDiseases

Loading Abstract...

Abstract

This study examines the effectiveness of ground penetrating radar (GPR) in detecting and categorizingdiseases hidden beneath urban road surfaces. Convolutional neural networks (CNNs), specifically transfer learning using AlexNet, enhance classification accuracy of cavity shapes to 90.5%. Even with limited data, this strategy performs better than current ones. An examination of subsurface cavity data from 1965 reveals four main types: water-rich bodies, hollow bodies, empty bodies, and loose bodies. These classifications are critical for assessing road hazards including sinkholes and voids. The precise interpretation of image features associated with cavity disorders remains challenging. This is due to several factors, including geological, human, and environmental impacts. Accurately assessing and identifying picture elements associated with dental issues remains challenging. Many additional variables can also contribute to subterranean diseases and cavities, including as environmental and geological causes, the physical and chemical characteristics of geotechnical materials, engineering activities carried out by humans, and the impact on the population or economics of the subterranean community

Loading PDF...

Loading Statistics...

Loading Video...