Abstract The CrackNet, an efficient architecture based on the Convolutional Neural Network (CNN), is proposed in this article for automated pavement crack detection 3D asphalt surfaces with explicit objective of pixel‐perfect accuracy. Unlike commonly used CNN, CrackNet does not have any pooling layers which downsize outputs previous layers. fundamentally ensures accuracy using newly developed technique invariant image width and height through all consists five includes more than one million...

A few recent developments have demonstrated that deep-learning-based solutions can outperform traditional algorithms for automated pavement crack detection. In this paper, an efficient deep network called CrackNet-V is proposed pixel-level detection on 3D asphalt images. Compared with the original CrackNet, has a deeper architecture but fewer parameters, resulting in improved accuracy and computation efficiency. Inspired by uses invariant spatial size through all layers such supervised...

Abstract A recurrent neural network (RNN) called CrackNet‐R is proposed in the article for fully automated pixel‐level crack detection on three‐dimensional (3D) asphalt pavement surfaces. In article, a new unit, gated multilayer perceptron (GRMLP), to recursively update internal memory of CrackNet‐R. Unlike widely used long short‐term (LSTM) and unit (GRU), GRMLP intended deeper abstractions inputs hidden states by conducting nonlinear transforms at gating units. implements two‐phase...

CrackNet is the result of an 18-month collaboration within a 10-person team to develop deep learning–based pavement crack detection software that demonstrated successes in terms consistency for both precision and bias. This paper proposes improved architecture called II enhanced learning capability faster performance. The proposed represents two major modifications on original CrackNet. First, feature generator, which provides handcrafted features through fixed nonlearnable procedures, no...

The classification of pavement crack heavily relies on the engineers' experience or hand-crafted algorithms. Convolutional Neural Network (CNN) has demonstrated to be useful for image classification, which provides an alternative traditional imaging This paper proposes a novel method using deep CNN automatically classify patches cropped from 3D images. In all, four supervised CNNs with different sizes receptive field are successfully trained. experimental results demonstrate that all...

Abstract Cracks are the most common damage type on pavement surface. Usually, cracks, especially small difficult to be accurately identified due background interference. Accurate and fast automatic road crack detection play a vital role in assessing conditions. Thus, this paper proposes an efficient lightweight encoder–decoder network for automatically detecting cracks at pixel level. Taking advantage of novel architecture integrating new hybrid attention blocks residual (RBs), proposed can...

Traffic sign recognition is critical for advanced driver assistant system and road infrastructure survey. Traditional traffic algorithms can't efficiently recognize signs due to its limitation, yet deep learning-based technique requires huge amount of training data before use, which time consuming labor intensive. In this study, transfer method introduced classification, significantly reduces the alleviates computation expense using Inception-v3 model. our experiment, Belgium Sign Database...

Abstract Simultaneous pixel‐level detection of multiple distresses and surface design features on complex asphalt pavements is a critical challenge in intelligent pavement survey. This paper proposes deep‐learning model named ShuttleNet to provide an efficient solution for this by implementing robust semantic segmentation pavements. The proposed aims at repeating the encoding–decoding round freely or even endlessly such that contexts different resolution levels can be learned integrated many...

Edge detection is an alternative method in the process for identifying and classifying pavement cracks automated evaluation systems. A number of edge detectors are widely used image processing; most specify only a spatial scale detecting edges. However, surface images frequently have various details at different scales. Therefore, wavelet-based multiscale technique can be candidate to extract information from images. Instead edges space domain, wavelet analysis has ability describe both...

There have been rapid developments in automated surveying of cracking pavements recent years. Laser-imaging technology has made the acquisition shadow-free images feasible. However, because complexity pavement surfaces, diverse characteristics cracks, presence foreign objects, and varying identification protocols, results recognition had limited use. A matched filtering algorithm is introduced for detection cracking. Unlike traditional edge approaches that adopt first-or second-order...

Skid resistance depends on the macro- and micro-textural characteristics of pavement surface. Mean profile depth (MPD) is a widely used two-dimensional macro-texture indicator calculated from single surface profile, while micro-texture primarily affected by aggregates contained within In this study, twenty-two sites, constructed using six common types preventive treatments eight typical sources in Oklahoma, are selected as field test beds. Pavement skid data collected parallel at highway...

Abstract Crack is one of the most important pavement condition indicators that are immediately relevant to water ingress and deterioration. In practices management, crack width has been extensively referenced by highway agencies determine severity. Accurate measurement meaningful for in understanding mechanism formation, predicting propagation. This article presents a new automatic method measuring using binary map images. The proposed introduces definition formulates it Laplace's Equation...

Abstract It is challenging to collect 3D pavement images with desired resolution for accurate texture measurement at driving speeds current devices, particularly in the longitudinal direction. This paper presents a novel superresolution technique recursive generative adversarial network, called Pavement Texture Super Resolution Generative Adversarial Network (PT‐SRGAN), reconstruct 0.1‐mm image from low‐resolution data faster measurement. With proposed pseudo‐Laplacian pyramid and an array...

Accurate identification of cracks at the pixel level on intricate asphalt pavements represents a crucial challenge in domain intelligent pavement assessment. The current advanced deep-learning networks encounter limitations simultaneously capturing both global context and local features cracks, leading to discontinuous segmentation results suboptimal recovery details. This paper proposes robust architecture named Mix-Graph CrackNet present an efficacious solution for this challenge....

The categorization and quantification of the type, severity, extent surface distress is a primary method for assessing condition highway pavements. Various methodologies were developed to automate pavement surveys. Significant technical advances made during past years. implementations image capturing subsystems include conventional analog-based area-scan, analog digital line-scan, laser scanning, shadow Moire method. Newer processing artificial neural net parallel processing. However,...

Since the AASHO road test of 1962, tremendous efforts have been devoted to improve methodologies pavement performance prediction. The successful implementation Network Optimization System (NOS) in Arizona Department Transportation (ADOT), 1980s, represented an advancement prediction methodology by using Markov‐process‐based transition‐probability matrices (TPMs) define transition process conditions. This paper addresses some inadequacies original NOS model. Two approaches were used evaluate...

Pavement distress survey plays an important role in pavement maintenance and management. distresses include items such as cracks identifiable with 2D images well potholes 3D images. Most of the current technologies applied system imaging 1D profiler. Information on characteristics is not used routine surveys, primarily due to technical difficulties hardware acquisition software algorithms. As result, most surveys for pavements are automated. A novel idea proposed paper apply stereovision...

Due to the complexity and diversity of pavement surfaces, cracking detection is a challenging task even for human operators. The automation generally requires robust algorithms with high level intelligence. From such perspective, Deep Learning, promising branch Artificial Intelligence, can serve as an advanced approach intelligence by learning from huge amount historical data enhancing capability behaving correctly under unforeseen complex environments. This paper proposes convolutional...

It is widely recognized that pavement surface texture important for friction and roadway safety. Mean profile depth, the primary index used to characterize macrotexture, a two-dimensional height indicator calculated from single profile, which insufficient represent texture, especially studies. This paper explores five categories of three-dimensional (3-D) areal parameters attributes develops relationship between 3-D parameters. The newly constructed Long-Term Pavement Performance Specific...

In this paper, a quantum state restoration scheme is proposed based on the environment-assisted error correction (EAEC) scheme. By introducing weak measurement reversal (WMR) operation, we will show how to recover an initial of open system without invoking random unitary decompositions which are known be absent in many important physical systems. We illustrate our and compare it with purely WMR operation case dissipative channel. scheme, successful probability for recovering unknown can...

Pavement cracking is a significant symptom of pavement deterioration and deficiency. Conventional manual inspections road condition are gradually replaced by novel automated inspection systems. As result, great amount surface information digitized these systems with high resolution. With data, cracks can be detected using crack detection algorithms. In this paper, fully algorithm for segmenting enhancing proposed, which consists four major procedures. First, preprocessing procedure employed...