With the advancement of machine learning (ML) and deep (DL), there is a great opportunity to enhance development automatic crack detection algorithms. In this paper, authors organize provide up-to-date information on ML-based algorithms for researchers more efficiently seek potential focus direction. The first reviewed 68 methods identify current trend development, pixel-level segmentation. then conducted performance evaluation 8 segmentation models using consistent metrics three-dimensional...

Image segmentation is the crucial step in automatic image distress detection and classification (e.g., types severities) has important applications for crack sealing. Although many researchers have developed pavement recognition algorithms, full automation remained a challenge. This first paper that uses scoring measure to quantitatively objectively evaluate performance of six different algorithms. Up-to-date research on comprehensively reviewed identify need. Six methods are then tested...

After decades of research and development, a fully automated system for pavement crack detection with an intensity-based two-dimensional (2D) imaging data acquisition under different lighting low intensity contrast conditions still remains challenge. With the advances sensor technology, three-dimensional (3D) laser technology that can collect high-resolution 3D continuous transverse profiles detecting cracks on basis their elevation rather than 2D has become available. This study, sponsored...

Potholes are one of the roadway distresses that negatively impact safety. With emerging sensing technology, three-dimensional (3D) pavement data, derived using 3D laser have become available for detecting cracking and rutting. This paper presents a pothole detection method data watershed method. Tests collected on 10th Street, Atlanta, Georgia 6 mi U.S. 80, Savannah, Georgia, has shown 94.97% accuracy, 90.80% precision, 98.75% recall. It been demonstrated proposed is promising can provide...

Single image dehazing is the ill-posed two-dimensional signal reconstruction problem. Recently, deep convolutional neural networks (CNN) have been successfully used in many computer vision problems. In this paper, we propose a Y-net that named for its structure. This network reconstructs clear images by aggregating multi-scale features maps. Additionally, Wavelet Structure SIMilarity (W-SSIM) loss function training step. proposed function, discrete wavelet transforms are applied repeatedly...

Existing state-of-the-art minimal path techniques work well to extract simple open curves in images when both endpoints of the curve are given as user input or one is and total length known advance. Curves which branch require even further prior from user, namely, each endpoint. In this work, we present a novel path-based algorithm works on much more general topologies with far fewer demands for initial compared algorithms. The two key novelties benefits new approach that 1) it may be used...

Automatic pavement crack segmentation has gained attention among researchers and transportation agencies over the past two decades. However, most existing algorithms using two-dimensional (2D) intensity images cannot provide a satisfactory performance. With advent of sensing technology, three-dimensional (3D) line laser imaging systems have become available. The objective this paper is to propose an enhanced dynamic optimization algorithm employing advantages 3D data improve segmentation....

Horizontal curves play a critical role in roadway safety by providing smooth transition between tangent sections. Because radii of horizontal are one the most fundamental elements geometry design, transportation agencies, e.g., state DOTs, need to measure them support network-level analysis. However, traditional methods that commonly used plan sheet reading method and chord-offset method, time consuming, labor intensive, inaccurate. Although some semiautomatic automatic have been developed...

Traffic signs are important roadway assets that provide critical guidance, including regulations and safety-related information, to road users. need be inventoried by transportation agencies. However, the traditional manual methods carried out in field dangerous, labor-intensive, time-consuming. There is a develop alternative cost-effectively inventory traffic signs. The research reported this paper, sponsored U.S. DOT Research Innovative Technology Administration Program, critically assess...

Abstract The detailed monitoring of jointed plain concrete pavement (JPCP) slab condition is essential for cost‐effective JPCP maintenance and rehabilitation. However, existing visual inspection practices classification are time‐consuming labor‐intensive. In this paper, we proposed an automated model based on convolutional neural networks (ConvNets), which the first to perform multi‐label both crack types severity levels. To handle different scales between states, includes a novel global...

Accurate pavement performance forecasting is critical in supporting transportation agencies’ predictive maintenance strategies: programs that prolong service life while using fewer resources. However, because of the complex nature deterioration, high accuracy for long-term and project-level challenging to traditional models. Therefore, researchers have taken advantage machine learning (ML) technology create more sophisticated models recent years. there are no extant studies compare different...

Abstract: Horizontal roadway curvature data are essential for safety analysis. However, collecting such is time-consuming, costly, and dangerous using traditional, manual surveying methods. It especially difficult to perform measurement when roadways have high traffic volumes. Thus, it would be valuable transportation agencies if could computed from photographic images taken low-cost digital cameras. This the first article that develops an algorithm emerging vision technology acquire...

High-resolution 2D intensity and 3D range data are available to support pavement crack detection classification. However, classification based on actual distress protocols specified by different state departments of transportation (DOTs) still remains a challenge because the complexity real-world In addition, developing performance measurements that can be used consistently among DOTs is another challenge. The objective this paper propose novel fundamental element (CFE) model multiscale...

An effective cross slope facilitates drainage on highways and prevents hydroplaning. There is a need for transportation agencies to identify measure road sections that have noneffective slopes so timely corrective maintenance can be performed. However, the traditional manual methods used by with digital level are time-consuming labor intensive; these not feasible conducting network-level cross-slope measurement. A proposed mobile measurement method uses emerging lidar technology accurately...

The sidewalk is an indispensable infrastructure for pedestrians, especially wheelchair users. Wheelchair users rely on quality sidewalks to facilitate safe and uninterrupted trips in their everyday lives. Transportation agencies are required evaluate regulatory compliance with the Americans Disabilities Act (ADA) responsible timely maintenance of inadequate sidewalks. However, these evaluation activities usually lacking because labor-intensive cost-prohibitive data collection process current...

Three-dimensional (3D) laser scanners have become a mainstream technology for the automatic assessment of pavement condition. The objective this study is to leverage highly accurate 3D data train supervised machine learning models condition estimation using low-cost vehicle-mounted smartphone sensor data. First, and were registered on common geographic information system (GIS) model road network. Second, recurrent neural networks (RNNs) with long short-term memory (LSTM) units trained...

Raveling is one of the most common asphalt pavement distresses. The survey its condition required for transportation agencies to ensure roadway safety and appropriately apply preservation rehabilitation treatments. However, traditional raveling survey, including determination severity, typically manually conducted by in-field visual inspection methods that are time consuming, labor intensive, error prone. Although automated detection severity classification models have been developed, these...

The hundreds of traffic sign types on the road and their various shapes colors make it difficult to develop a generalized method detection. Consequently, agencies performing inventory must manually review millions roadway video log images. This paper proposes an innovative image processing model that automatically detects signs dramatically reduces workload. In test proposed using 37,640 images provided by Louisiana Department Transportation Development, 86 percent manual efforts can be...

Faulting has traditionally been collected by using manual methods, which are labor intensive, time-consuming, and hazardous to workers drivers. Therefore, alternative methods for effectively safely collecting faulting data needed. With emerging laser technology originally designed crack detection, high-resolution, full lane-width coverage, three-dimensional (3D) continuous pavement profile can now be acquired. This paper critically assesses the feasibility of this 3D measuring with a special...

Automated pavement crack detection is essential to a cost-effective asset management system. Many automated algorithms (CDAs) have been developed, but they lack standardized performance evaluation system, which urgently needed. This paper presents comprehensive, quantitative CDA system (CDA-PES) that includes: (1) an enhanced Hausdorff distance–based method; (2) consistent data set designed with diverse types and conditions affect performance; (3) multilevel, categorized, scoring reporting...