检索结果-内蒙古大学图书馆

Deep learning-based autonomous concrete crack evaluation through hybrid image scanning

STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL 2019年第5-6期18卷 1722-1737页

作者： Jang, Keunyoung Kim, Namgyu An, Yun-Kyu Sejong Univ Dept Architectural Engn 209 Neungdong Ro Seoul 143747 South Korea Sejong Univ Dept Civil & Environm Engn Seoul South Korea

This article proposes a deep learning-based autonomous concrete crack detection technique using hybrid images. The hybrid images combining vision and infrared thermography images are able to improve crack detectability while minimizing false alarms. In particular, large-scale concrete-made infrastructures such as bridge and dam can be effectively inspected by spatially scanning the unmanned vehicle-mounted hybrid imaging system including a vision camera, an infrared camera, and a continuous-wave line laser. However, the expert-dependent decision-making for crack identification which has been widely used in industrial fields is often cumbersome, time-consuming, and unreliable. As a target concrete structure gets larger, automated decision-making becomes more desirable from the practical point of view. The proposed technique is able to achieve automated crack identification and visualization by transfer learning of a well-trained deep convolutional neural network, that is, GoogLeNet, while retaining the advantages of the hybrid images. The proposed technique is experimentally validated using a lab-scale concrete specimen with cracks of various sizes. The test results reveal that macro- and microcracks are automatically visualized while minimizing false alarms.

关键词： concrete crack detection deep convolutional neural network hybrid image scanning vision image infrared thermography structural health monitoring

来源：评论

学校读者我要写书评

暂无评论

Performance Degradation Assessment of concrete Beams Based on Acoustic Emission Burst Features and Mahalanobis-Taguchi System

引用

SENSORS 2020年第12期20卷 3402.-3402.页

作者： Habib, Md Arafat Rai, Akhand Kim, Jong-Myon Univ Ulsan Sch Comp Engn & Informat Technol 93 Daehak Ro Ulsan 44610 South Korea

Acoustic emission (AE) has been used extensively for structural health monitoring based on the stress waves generated due to evolution of cracks in concrete structures. A major concern while using AE features is that each of them responds differently to the fractures in concrete structures. To tackle this problem, Mahalanobis-Taguchi system (MTS) is utilized, which fuses the AE feature space to provide comprehensive and reliable degradation indicator with a feature selection method to determine useful features. Further, majority of the existing investigations gave little attention to naturally occurring cracks, which are actually more difficult to detect. In this study, a novel degradation indicator (DI) based on AE features and MTS is proposed to indicate the performance degradation in reinforced concrete beams. The experimental results confirm that the MTS can successfully distinguish between healthy and faulty conditions. To alleviate the noise from the DI obtained through MTS, a noise-removal strategy based on Chebyshev inequality is suggested. The results show that the proposed DI based on AE features and MTS is capable of detecting early stage cracks as well as development of damage in concrete beams.

关键词： concrete crack detection acoustic emission features Mahalanobis-Taguchi system

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Neural Networks for sUAS-Assisted Structural Inspections: Feasibility and Application

Deep Learning Neural Networks for sUAS-Assisted Structural I...

引用

INTERNATIONAL CONFERENCE ON UNMANNED AIRCRAFT SYSTEMS (ICUAS)

作者： Dorafshan, Sattar Coopmans, Calvin Thomas, Robert J. Maguire, Marc Utah State Univ Dept Civil & Environm Engn Logan UT 84322 USA Utah State Univ Utah Water Res Lab AggieAir Logan UT 84322 USA

ISBN: (纸本)9781538613542

This paper investigates the feasibility of using a Deep Learning Convolutional Neural Network (DLCNN) in inspection of concrete decks and buildings using small Unmanned Aerial Systems (sUAS). The training dataset consists of images of lab-made bridge decks taken with a point-and-shoot high resolution camera. The network is trained on this dataset in two modes: fully trained (94.7% validation accuracy) and transfer learning (97.1% validation accuracy). The testing datasets consist of 1620 sub-images from bridge decks with the same cracks, 2340 sub-images from bridge decks with similar cracks, and 3600 sub-images from a building with different cracks, all taken by sUAS. The sUAS used in the first dataset has a low-resolution camera whereas the sUAS used in the second and third datasets has a camera comparable to the point-and-shoot camera. In this study it has been shown that it is feasible to apply DLCNNs in autonomous civil structural inspections with comparable results to human inspectors when using off-the-shelf sUAS and training datasets collected with point-and-shoot handheld cameras.

关键词： small Unmanned Aerial Systems Deep Learning Convolutional Neural Networks concrete crack detection Infrastructure Inspection Remote Sensing

来源：评论

学校读者我要写书评

暂无评论

crack detection Based on Support Vector Data Description 29

Crack Detection Based on Support Vector Data Description

引用

第29届中国控制与决策会议

作者： LIN Weiguo LIN Yaru Wang Fang College of Information Science and Technology Beijing University of Chemical Technology

ISBN: (纸本)9781509046584

It is difficult to detect concrete cracks because of the existence of background *** solve this problem,some methods for crack detection on concrete surfaces are *** to shape features,a new concrete crack detection method is *** of all,iteration method is applied to get the optimal threshold for image segmentation after grayscale ***,binary images are processed by morphological closing operation and *** including eccentricity,circularity and packing density are selected as input training vectors for Support Vector Data Description(SVDD).The experimental results show that crack detection method based on SVDD can accurately distinguish cracks from other kinds of defects(non-crack) and reduce the false negative detections.

关键词： concrete crack detection Optimal threshold Shape feature Support Vector Data Description

来源：评论

学校读者我要写书评

暂无评论

Initial Proof of Concept on the Use of 3D Models to Identify Structural Deviations of Bridge Components

Initial Proof of Concept on the Use of 3D Models to Identify...

引用

IEEE SoutheastCon

作者： Luis Daniel Otero Mark Moyou Adrian Peter Dennis Dalli Nick Gagliardo Department of Engineering Systems Florida Institute of Technology Melbourne FL USA

ISBN: (纸本)9781538661345

This research paper presents ongoing research work towards developing an effective remote sensing system for bridge inspections. The paper describes various tests that were conducted to address a research question regarding the applicability of 3D models developed from mobile LiDAR data to help identify signs of bridge settlement. The general solution approach involved test planning, collecting data, aligning LiDAR data into a common coordinate system, developing 3D models, and evaluating the use of the resulting models to identify structural deviations. A mockup bridge structure was developed using PVC material to conduct initial experiments prior to conducting more expensive field tests. Prior to data collection, the LiDAR sensor was placed on a tripod mount at a distance of x feet from the PVC structure, and a portion of the PVC structure was inclined y inches using 0.5 inches thick wooden stakes. The resulting 3D models showed that a 0.5 inch inclination was easily identified with the sensor located 10 feet from the structure, and using only 20 LiDAR scans. Future areas of research are discussed to extend the ongoing work presented in this paper.

关键词： Structural inspections Bridge inspections concrete crack detection Transportation systems engineering

来源：评论

学校读者我要写书评

暂无评论

Temperature measurement and damage detection in concrete beams exposed to fire using PPP-BOTDA based fiber optic sensors

引用

SMART MATERIALS AND STRUCTURES 2017年第10期26卷 105034页

作者： Bao, Yi Hoehler, Matthew S. Smith, Christopher M. Bundy, Matthew Chen, Genda Missouri Univ Sci & Technol Dept Civil Architectural & Environm Engn 1401 N Pine St Rolla MO 65409 USA NIST Natl Fire Res Lab 100 Bur Dr Gaithersburg MD 20899 USA

In this study, Brillouin scattering-based distributed fiber optic sensor is implemented to measure temperature distributions and detect cracks in concrete structures subjected to fire for the first time. A telecommunication-grade optical fiber is characterized as a high temperature sensor with pulse pre-pump Brillouin optical time domain analysis (PPP-BODTA), and implemented to measure spatially-distributed temperatures in reinforced concrete beams in fire. Four beams were tested to failure in a natural gas fueled compartment fire, each instrumented with one fused silica, single-mode optical fiber as a distributed sensor and four thermocouples. Prior to concrete cracking, the distributed temperature was validated at locations of the thermocouples by a relative difference of less than 9%. The cracks in concrete can be identified as sharp peaks in the temperature distribution since the cracks are locally filled with hot air. concrete cracking did not affect the sensitivity of the distributed sensor but concrete spalling broke the optical fiber loop required for PPP-BOTDA measurements.

关键词： concrete crack detection distributed fiber optic sensor PPP-BOTDA structural fire safety temperature distribution

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Neural Networks for sUAS-Assisted Structural Inspections: Feasibility and Application

Deep Learning Neural Networks for sUAS-Assisted Structural I...

引用

International Conference on Unmanned Aircraft Systems

作者： Sattar Dorafshan Robert J. Thomas Calvin Coopmans Marc Maguire Dept. of Civil and Environmental Engineering Utah State University Logan UT USA AggieAir at Utah Water Research Laboratory Utah State University Logan UT USA

关键词： Small Unmanned Aerial Systems Deep Learning Convolutional Neural Networks concrete crack detection Infrastructure Inspection Remote Sensing

来源：评论

学校读者我要写书评

暂无评论

crack identification for reinforced concrete using PZT based smart rebar active sensing diagnostic network

Crack identification for reinforced concrete using PZT based...

引用

Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, and Civil Infrastructure 2016

作者： Song, N. N. Wu, F. Shanghai Jiao Tong Univ Sch Naval Architecture Ocean & Civil Engn Dept Civil Engn Shanghai Peoples R China

ISBN: (纸本)9781510600454

An active sensing diagnostic system using PZT based smart rebar for SHM of RC structure has been currently under investigation. Previous test results showed that the system could detect the de-bond of concrete from reinforcement, and the diagnostic signals were increased exponentially with the de-bonding size. Previous study also showed that the smart rebar could function well like regular reinforcement to undertake tension stresses. In this study, a smart rebar network has been used to detect the crack damage of concrete based on guided waves. Experimental test has been carried out for the study. In the test, concrete beams with 2 reinforcements have been built. 8 sets of PZT elements were mounted onto the reinforcement bars in an optimized way to form an active sensing diagnostic system. A 90 kHz 5-cycle Hanning-windowed tone burst was used as input. Multiple cracks have been generated on the concrete structures. Through the guided bulk waves propagating in the structures from actuators and sensors mounted from different bars, crack damage could be detected clearly. Cases for both single and multiple cracks were tested. Different crack depths from the surface and different crack numbers have been studied. Test result shows that the amplitude of sensor output signals is deceased linearly with a propagating crack, and is decreased exponentially with increased crack numbers. From the study, the active sensing diagnostic system using PZT based smart rebar network shows a promising way to provide concrete crack damage information through the "talk" among sensors.

关键词： PZT smart rebar net work Guided wave concrete crack detection experimental test

来源：评论

学校读者我要写书评

暂无评论

Optical crack following on tunnel surfaces

Optical crack following on tunnel surfaces

引用

Conference on Two- and Three-Dimensional Methods for Inspection and Metrology IV

作者： Paar, Gerhard Caballo-Perucha, Maria d. P. Kontrus, Heiner Sidla, Oliver Joanneum Res Inst Digital Image Proc Wastiangasse 6 A-8010 Graz Austria Dibit Messtech GmbH A-6020 Innsbruck Austria

ISBN: (纸本)9780819464804

One of the most important monitoring tasks of tunnel inspection is the observation of cracks. This paper describes an approach for crack following using mid-resolution (2-5mm per pixel) images of the tunnel surface. A mosaic on the basis of the tunnel design surface is built from images taken with a mobile platform. On this image representing the unwrapped tunnel surface texture the starting points of each crack are found semiautomatically using a modified Hough transform. crack following takes place on the basis of local line fitting and exhaustive search in both directions of the crack, taking into account several restrictions, rules and optimization criteria to find the correct crack trajectory. A practical implementation polygonizes the extracted cracks and feeds them into a tunnel inspection data base. The method is applicable to various types of background texture as expected in the tunnel environment.

关键词： tunnel inspection crack following concrete crack detection line following

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：