Structural inspection and maintenance of bridges are essential to improve the safety and sustainability of the infrastructure systems. Visual inspection using non-equipped eyes is the principal method of detecting surface defects of bridges, which is time-consuming, unsafe, and encounters inspectors falling risks. Therefore, there is a need for automated bridge inspection. Recently, Light Detection and Ranging (LiDAR) scanners are used for detecting surface defects. LiDAR scanners can collect high-quality 3D point cloud datasets. In order to automate the process of structural inspection, it is important to collect proper datasets and use an efficient approach to analyze them and find the defects. Deep Neural Networks (DNNs) have been recently used for detecting 3D objects within 3D point clouds. PointNet and PointNet++ are deep neural networks for classification, part segmentation, and semantic segmentation of point clouds that are modified and adapted in this work to detect surface concrete defects. The research contributions are: (1) Designing a LiDAR-equipped UAV platform for structural inspection using an affordable 2D scanner for data collection, and (2) Proposing a method for detecting concrete surface defects using deep neural networks based on LiDAR generated point clouds. Training and testing datasets are collected from four concrete bridges in Montréal and annotated manually. The point cloud dataset prepared in five areas, which contain more than 51 million points and 2,572 annotated defects in four classes of crack, light spalling, medium spalling, and severe spalling. The accuracies of 75% (adapted PointNet) and 79% (adapted PointNet++) in detecting defects are achieved in binary semantic segmentation.

" TRB's second Strategic Highway Research Program (SHRP 2) Report S2-R06A-RR-1: Nondestructive Testing to Identify Concrete Bridge Deck Deterioration identifies nondestructive testing technologies for detecting and characterizing common forms of deterioration in concrete bridge decks.The report also documents the validation of promising technologies, and grades and ranks the technologies based on results of the validations.The main product of this project will be an electronic repository for practitioners, known as the NDToolbox, which will provide information regarding recommended technologies for the detection of a particular deterioration. " -- publisher's description.

Bridge decks deteriorate over time as a result of deicing salts, freezing-and-thawing, and heavy use, resulting in internal defects. According to a 2006 study by the American Society of Civil Engineers, 29% of bridges in the United States are considered structurally deficient or functionally obsolete. Ground penetrating radar (GPR) is a promising non-destructive evaluation technique for assessing subsurface conditions of bridge decks. However, the analysis of GPR scans is typically done manually, where the accuracy of the detection process depends on the technician's trained eye. In this work, a framework is developed to automate the detection, locailzation, and characterization of subsurface defects inside bridge decks. This framework is composed of a fractal-based feature extraction algorithm to detect defective regions, a deconvolution algorithm using banded-ICA to reduce overlapping between reflections and to estimate the depth of defects, and a classification algorithm using principal component analysis to identify main features in defective regions. This framework is implemented and simulated using MATLAB and GPR real scans of simulated concrete bridge decks. This framework, as demonstrated by the experimental results, has the following contributions to the current body of knowledge in ground penetrating radar detection and analysis techniques, and in concrete bridge deck condition assessment: 1) developed a framework that integrated detection, localization, and classificationof subsurface defects inside concrete bridge decks, 2) presented a comparison between the most common fractal methods to determine the most suitable one for bridge deck condition assessment, 3) introduced a fractal-based feature extraction algorithm that is capable of detecting and horizontally labeling defective regions using only the underlying GPR B-scan without the need for a training dataset, 4) developed a deconvolution algorithm using EFICA to detect embedded defects in bridge decks, 5) introduced an automated identification methodology of defective regions which can be integrated into a CAD system that allows for better visual assessment by the maintenance engineer and has the potential to eliminate human interpretation errors and reduce condition assessment time and cost, and 6) presented an investigation and a successful attempt to classify some of the common defects in bridge decks.

Structural health monitoring of concrete bridges requires accurate and efficient surface crack detection. Early detection of cracks helps prevent further damage. Safety inspection tests are conducted at regular intervals to assess deterioration. Traditional methods involve detection of cracks by human visual inspection. These methods are costly, inefficient and labor intensive, especially for long-span bridges. This thesis presents the use of computer vision and pattern recognition techniques in assessment of cracks on a concrete bridge surface. Bridge deck images are first collected using high-resolution cameras mounted on a robot. Statistical inference algorithms are then implemented to build an automated crack detection system. The proposed machine learning method reduces manual effort and enables automatic labeling over large bridge deck areas to quantify size and location for future reference or comparisons. A panoramic camera is used for the purpose of context localization. Additionally, we demonstrate image-stitching to obtain a coherent spatial mosaic of the bridge deck.

Bridge management system (BMS) is an effective mean for managing bridges throughout their design life. BMS requires accurate collection of data pertinent to bridge conditions. Non Destructive Evaluation methods (NDE) are automated accurate tools used in BMS to supplement visual inspection. This research provides overview of current practices in bridge inspection and in-depth study of thirteen NDE methods for condition assessment of concrete bridges and eleven for structural steel bridges. The unique characteristics, advantages and limitations of each method are identified along with feedback on their use in practice. Comparative study of current practices in bridge condition rating, with emphasis on the United States and Canada is also performed. The study includes 4 main criteria: inspection levels, inspection principles, inspection frequencies and numerical ratings for 4 provinces and states in North America and 5 countries outside North America. Considerable work has been carried out using a number of sensing technologies for condition assessment of civil infrastructure. Fewer efforts, however, have been directed for integrating the use of these technologies. This research presents a newly developed method for automated condition assessment and rating of concrete bridge decks. The method integrates the use of ground penetrating radar (GPR) and infrared thermography (IR) technologies. It utilizes data fusion at pixel and feature levels to improve the accuracy of detecting defects and, accordingly, that of condition assessment. Dynamic Bayesian Network (DBN) is utilized at the decision level of data fusion to overcome cited limitations of Markov chain type models in predicting bridge conditions based on prior inspection results. Pixel level image fusion is applied to assess the condition of a bridge deck in Montreal, Canada using GPR and IR inspection results. GPR data are displayed as 3D from 24 scans equally spaced by 0.33m to interpret a section of the bridge deck surface. The GPR data is fused with IR images using wavelet transform technique. Four scenarios based on image processing are studied and their application before and after data fusion is assessed in relation to accuracy of the employed fusion process. Analysis of the results showed that bridge condition assessment can be improved with image fusion and, accordingly, support inspectors in interpretation of the results obtained. The results also indicate that predicted bridge deck condition using the developed method is very close to the actual condition assessment and rating reported by independent inspection. The developed method was also applied and validated using three case studies of reinforced concrete bridge decks. Data and measurements of multiple NDE methods are extracted from Iowa, Highway research board project, 2011. The method utilizes data collected from ground penetrating radar (GPR), impact echo (IE), Half-cell potential (HCP) and electrical resistivity (ER). The analysis results of the three cases indicate that each level of data fusion has its unique advantage. The power of pixel level fusion lies in combining the location of bridge deck deterioration in one map as it appears in the fused image. While, feature fusion works in identification of specific types of defects, such as corrosion, delamination and deterioration. The main findings of this research recommend utilization of data fusion within two levels as a new method to facilitate and enhance the capabilities of inspectors in interpretation of the results obtained. To demonstrate the use of the developed method and its model at the decision level of data fusion an additional case study of a bridge deck in New Jersey, USA is selected. Measurements of NDE methods for years 2008 and 2013 for that bridge deck are used as input to the developed method. The developed method is expected to improve current practice in forecasting bridge deck deterioration and in estimating the frequency of inspection. The results generated from the developed method demonstrate its comprehensive and relatively more accurate diagnostics of defects.