4.7 Article

Development of a Concrete Floating and Delamination Detection System Using Infrared Thermography

期刊

IEEE-ASME TRANSACTIONS ON MECHATRONICS
卷 26, 期 6, 页码 2835-2844

出版社

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TMECH.2021.3106867

关键词

Delamination; Concrete; Cameras; Inspection; Bridges; Temperature measurement; Servers; Boosting; civil engineering; concrete; infrared surveillance; inspection; machine learning

向作者/读者索取更多资源

This study developed a system for inspecting concrete structures using infrared thermography, which overcomes the challenges of traditional inspection methods and utilizes machine learning technology to automatically detect temperature irregularities, offering an effective solution for preventing concrete spalling.
Spalling of concrete fragments due to the deterioration of concrete structures can cause property damage or serious and even fatal accidents; thus, there is a need to detect such deterioration. Generally, the hammering test is employed as the main inspection method to prevent such concrete spalling; however, it requires close contact with the structure being tested. Getting close to the structure for inspection is expensive and time consuming, and if the structure is high up, there is a risk of falling. Therefore, in this study, we developed a system for inspecting concrete structures without approaching them, using infrared thermography. In order to detect floating and delamination using infrared thermography, it is necessary to find temperature irregularities caused by such damage from an infrared image, but such an inspection method has not been realized so far. There are two main reasons for this. First, it is difficult to evaluate whether the concrete structure is in an appropriate temperature condition suitable for detecting the floating and delamination. Second, it is difficult to detect temperature irregularities caused by floating and delamination among the various causes of temperature irregularities. In this study, we resolved these issues by developing equipment to investigate whether the object is in an appropriate temperature condition for proper photography and by developing a machine learning-based method to automatically detect only the temperature irregularities caused by floating and delamination. By resolving these issues, we have developed a promising novel inspection method for the prevention of concrete spalling, which is reported in this article.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

Article Construction & Building Technology

Random forest-based evaluation technique for internal damage in reinforced concrete featuring multiple nondestructive testing results

Pang-jo Chun, Isao Ujike, Kohei Mishima, Masahiro Kusumoto, Shinichiro Okazaki

CONSTRUCTION AND BUILDING MATERIALS (2020)

Article Robotics

Innovative technologies for infrastructure construction and maintenance through collaborative robots based on an open design approach

Keiji Nagatani, Masato Abe, Koichi Osuka, Pang-jo Chun, Takayuki Okatani, Mayuko Nishio, Shota Chikushi, Takamitsu Matsubara, Yusuke Ikemoto, Hajime Asama

Summary: The socio-economic challenges facing the construction industry in Japan, such as the COVID19 pandemic, aging population, aging infrastructure, and severe natural disasters, are addressed by the project "Collaborative robots for adaptation of diverse environments and innovation of infrastructure construction." This project, part of the Moonshot project organized by the Japanese Cabinet Office, focuses on the open design approach for robotics, with research plans involving robots that adapt to the environment, multi-modal AI for environment evaluation, and physical AI with dynamic collaboration of multi-robots.

ADVANCED ROBOTICS (2021)

Article Robotics

Development of an excavator-avoidance system for buried pipes

Ryosuke Yajima, Shinya Katsuma, Makoto Suzuki, Fumiya Matsushita, Shunsuke Hamasaki, Pang-jo Chun, Keiji Nagatani, Genki Yamauchi, Takeshi Hashimoto, Atsushi Yamashita, Hajime Asama, Kazumasa Ozawa

Summary: A hydraulic excavator avoidance system was developed to automatically detect and avoid underground pipes, enabling safe and efficient excavation operations. Ground-penetrating radar and deep learning techniques were employed, along with autonomous excavation path generation, to avoid collisions with buried pipes.

ADVANCED ROBOTICS (2021)

Article Chemistry, Analytical

Study on Accuracy Improvement of Slope Failure Region Detection Using Mask R-CNN with Augmentation Method

Shiori Kubo, Tatsuro Yamane, Pang-jo Chun

Summary: This study proposes a deep learning-based semantic segmentation method to automatically detect slope failure regions, with the use of data augmentation to improve detection accuracy.

SENSORS (2022)

Article Green & Sustainable Science & Technology

Surrogate Model Development for Slope Stability Analysis Using Machine Learning

Xianfeng Li, Mayuko Nishio, Kentaro Sugawara, Shoji Iwanaga, Pang-jo Chun

Summary: This study implemented machine learning to analyze slope stability and developed a classification model and a regression model. The classification model achieved a testing accuracy of 0.9222, while the regression model had a high correlation coefficient R-square value of 0.9989 and a low test mean squared error value. Machine learning shows potential benefits in slope stability analysis.

SUSTAINABILITY (2023)

Article Chemistry, Multidisciplinary

Corrosion Damage Detection in Headrace Tunnel Using YOLOv7 with Continuous Wall Images

Shiori Kubo, Nobuhiro Nakayama, Sadanori Matsuda, Pang-jo Chun

Summary: This article discusses the deterioration of infrastructure built during Japan's high economic growth period and the need for maintenance and management of these structures. The aim of the study is to establish a system for investigating and assessing deformation points in headrace tunnel walls and to develop a system that automatically accumulates and plots damage locations and distributions.

APPLIED SCIENCES-BASEL (2023)

Article Engineering, Electrical & Electronic

ViTALnet: Anomaly on Industrial Textured Surfaces With Hybrid Transformer

Xian Tao, Chandranath Adak, Pang-Jo Chun, Shaohua Yan, Huaping Liu

Summary: The coexistence of subtle and long-range anomalies in real-world industrial applications brings challenges for anomaly localization. Existing methods do not consider learning local and global features simultaneously, which leads to inaccurate localization results. Therefore, a hybrid transformer model called ViTALnet is proposed, which leverages fine-grained feature reconstruction. ViTALnet first extracts local discriminatory features using a vision transformer (ViT) and then integrates global attention and a pyramidal architecture for fine-grain anomaly localization.

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2023)

Article Engineering, Civil

Difference in tensile load-carrying capacity of a corroded water pipeline due to circumferential location

Young Jun Nam, Pang-jo Chun, Cheol Ho Bae, Jeong Hyun Kim, Yun Mook Lim

Summary: This study focuses on accurately simulating the tensile performance of corroded specimens, taking into account the variations in mechanical properties along their circumference. By conducting experimental investigations and utilizing a 3D scanning system, the surface condition of the entire corroded pipeline was accurately represented in finite element models, resulting in a comprehensive comparison between the simulation and experiment results.

STRUCTURES (2023)

暂无数据