Pavement Surface Images Research Articles

Measuring the Macrotexture of Pavement Surface Using an Image Processing Technique

The objective of this study was to propose an image processing-based index for measuring pavement macrotexture at the network level. This index enables macrotexture to be measured easily and inexpensively using images collected at traffic speed. The study involved collecting pavement surface images at a constant traffic speed on a test section specifically designed and constructed for this purpose, with three surface course mixes that are commonly used in India, namely, bituminous concrete, stone matrix asphalt, and gap-graded rubberized bituminous mix. Additionally, macrotexture data with regard to mean texture depth (MTD) from the sand patch test and mean profile depth (MPD) from laser sensor-based measurements were obtained at the locations where the images were captured. The surface macrotexture index (SMI), which was derived from wavelet transform-based image texture analysis, was compared with the MTD and MPD data. The results showed that the SMI is an accurate indicator of pavement surface macrotexture. In addition, the study showcased the application of an unsupervised machine learning algorithm to identify and replace outliers in the SMI data that resulted from isolated spots with dirt, pavement markings, and wet surfaces. The research also established relationships between the proposed SMI and MTD/MPD. These relationships are reliable and can be used to predict the commonly used pavement surface construction quality measure MTD and the network-level skid resistance indicator MPD.

Classification Model Based on U-Net for Crack Detection from Asphalt Pavement Images

The purpose of our study is to detect cracks accurately from asphalt pavement surface images, which includes unexpected objects, non-uniform illumination, and irregularities in surfaces. We propose a method to construct a classification Convolutional Neural Network (CNN) model based on the pre-trained U-Net, which is a well-known semantic segmentation model. Firstly, we train the U-Net with a limited amount of the asphalt pavement surface dataset which is obtained by a Mobile Mapping System (MMS). Then, we use the encoder of the trained U-Net as a feature extractor to construct a classification model, and train by fine-tuning. We describe comparative evaluations with VGG11, ResNet18, and GoogLeNet as well-known models constructed by transfer learning using ImageNet, which is a large size dataset of natural images. Experimental results show our model has high classification performance, compared to the other models constructed by transfer learning using ImageNet. Our method is effective to construct convolutional neural network model using the limited training dataset.

Measurement of pavement rutting trajectories on two-lane highway using the 3D line scanning laser system

ABSTRACT Lane departure is one of the important reasons for traffic accidents on a two-lane highway. However, the vehicles’ trajectories obtained by the traditional methods only represent wheel wandering in specifical situations, but do not reflect a regular pattern of lane departure. Pavement rutting trajectories are applied to investigate the regular pattern of lane departure in this paper. A 3D line scanning laser system is utilized to obtain 3D deformed pavement surface images, and rutting trajectories are measured based on the collected data. Inertial Measurement Unit and Distance Measuring Instrument are used to describe lane centreline trajectory of two-lane highways, and curve radius and curve length are measured. A test site on a two-lane highway is chosen as the test bed. Based on the field data, the impacts of undivided opposite lanes, curve turning direction, geometric features, curve radius and curve length on lane departure are analyzed, and the hazardous locations with large lane departures are identified. The significance of this study is that it provides a method to describe a regular pattern of lane departure on two-lane highways, and also benefits transportation agencies in traffic safety analysis and the horizontal alignment design of two-lane highways.

Crack Detection and Classification in Moroccan Pavement Using Convolutional Neural Network

Crack is a condition indicator of the pavement’s structure. Generally, crack detection is an essential task for effective diagnosis of the road network. Moreover, evaluation of road quality is necessary to ensure traffic security. Since 2011, a periodic survey of approximately 57,500 km of Moroccan roads has been performed using an inspection vehicle (SMAC) which is equipped with high resolution cameras and GPS/DGPS receivers. Until recently, the teams of the National Center for Road Studies and Research (CNER) analyzed road surface states by visualization of pavement surface image sequences captured by the Multifunctional Pavement Assessment System (SMAC) in order to detect defects in road surfaces and classify them according to their type. However, this method involves manual processing and is complex, time consuming and subjective. In this paper, we propose an automated methodology for crack detection and classification in Moroccan flexible pavements using Convolutional Neural Networks (CNN). Transfer learning is also applied by testing a pre-trained Visual Geometry Group 19 (VGG-19) model. For the dataset used in this paper, the results indicate that good crack detection and classification are achieved using both models.

A potential crack region method to detect crack using image processing of multiple thresholding

In this paper, a potential crack region method is proposed to detect road pavement cracks by using the adaptive threshold. To reduce the noises of the image, the pre-treatment algorithm was applied according to the following steps: grayscale processing, histogram equalization, filtering traffic lane. From the image segmentation methods, the algorithm combines the global threshold and the local threshold to segment the image. According to the grayscale distribution characteristics of the crack image, the sliding window is used to obtain the window deviation, and then, the deviation image is segmented based on the maximum inter-class deviation. Obtain a potential crack region and then perform a local threshold-based segmentation algorithm. Real images of pavement surface were used at the Su Tong Li road in Suzhou, China. It was found that the proposed approach could give a more explicit description of pavement cracks in images. The method was tested on 509 images of the German asphalt pavement distress (Gap) dataset: The test results were found to be promising (precision = 0.82, recall = 0.81, F1 score = 0.83).

Development of a Deep Convolutional Neural Network for the Prediction of Pavement Roughness from 3D Images

Current roughness prediction models require extensive input data including pavement distress, climatic, and traffic data, which may be difficult to collect. In addition, these models have geographical limitations; therefore, a significant bias is expected if these models are used without recalibration. Pavement digital images can reflect both surface distresses and other surface irregularities that may affect pavement roughness conditions. In this study, convolutional neural networks (CNNs) were used to classify pavement sections into different roughness categories and to estimate International Roughness Index (IRI) values using pavement surface images. Furthermore, the effectiveness of artificial neural network (ANN) and multinomial logistic (MNL) regression models to categorize pavement sections into different roughness conditions was investigated. A pretrained CNN model was trained and validated using 850 three-dimensional (3D) pavement surface images, which were extracted from the Louisiana DOT and Development (LaDOTD) pavement management system (PMS) inventory. In addition, 1,142 test observations including IRI measurements and distress data were used to develop the ANN-based pattern recognition and MNL models. The developed CNN model outperformed the ANN and MNL models with an accuracy of 93.4% in the training stage. In addition, the CNN model predicted IRI values with a coefficient of determination (R2) of 0.985 and an average error of 5.9%.

Automated Detection and Classification of Pavement Distresses using 3D Pavement Surface Images and Deep Learning

Pavement distresses lead to pavement deterioration and failure. Accurate identification and classification of distresses helps agencies evaluate the condition of their pavement infrastructure and assists in decision-making processes on pavement maintenance and rehabilitation. The state of the art is automated pavement distress detection using vision-based methods. This study implements two deep learning techniques, Faster Region-based Convolutional Neural Networks (R-CNN) and You Only Look Once (YOLO) v3, for automated distress detection and classification of high resolution (1,800 × 1,200) three-dimensional (3D) asphalt and concrete pavement images. The training and validation dataset contained 625 images that included distresses manually annotated with bounding boxes representing the location and types of distresses and 798 no-distress images. Data augmentation was performed to enable more balanced representation of class labels and prevent overfitting. YOLO and Faster R-CNN achieved 89.8% and 89.6% accuracy respectively. Precision-recall curves were used to determine the average precision (AP), which is the area under the precision-recall curve. The AP values for YOLO and Faster R-CNN were 90.2% and 89.2% respectively, indicating strong performance for both models. Receiver operating characteristic (ROC) curves were also developed to determine the area under the curve, and the resulting area under the curve values of 0.96 for YOLO and 0.95 for Faster R-CNN also indicate robust performance. Finally, the models were evaluated by developing confusion matrices comparing our proposed model with manual quality assurance and quality control (QA/QC) results performed on automated pavement data. A very high level of match to manual QA/QC, namely 97.6% for YOLO and 96.9% for Faster R-CNN, suggest the proposed methodology has potential as a replacement for manual QA/QC.

Use of Deep Learning Image Classification Models and Vehicle Mounted Cameras for Automatic Pavement Pothole Detection

PURPOSES : This study uses deep learning image classification models and vehicle-mounted cameras to detect types of pavement distress — such as potholes, spalling, punch-outs, and patching damage — which require urgent maintenance. METHODS : For the automatic detection of pavement distress, the optimal mount location on a vehicle for a regular action camera was first determined. Using the orthogonal projection of obliquely captured surface images, morphological operations, and multi-blob image processing, candidate distressed pavement images were extracted from road surface images of a 16,036 km in-lane distance. Next, the distressed pavement images classified by experts were trained and tested for evaluation by three deep learning convolutional neural network (CNN) models: GoogLeNet, AlexNet, and VGGNet. The CNN models were image classification tools used to identify and extract the combined features of the target images via deep layers. Here, a data augmentation technique was applied to produce big distress data for training. Third, the dimensions of the detected distressed pavement patches were computed to estimate the quantity of repair materials needed. RESULTS : It was found that installing cameras 1.8 m above the ground on the exterior rear of the vehicle could provide clear pavement surface images with a resolution of 1 cm per pixel. The sensitivity analysis results of the trained GoogLeNet, AlexNet, and VGGNet models were 93 %, 86 %, and 72 %, respectively, compared to 62.7 % for the dimensional computation. Following readjustment of the image categories in the GoogLeNet model, distress detection sensitivity increased to 94.6 %. CONCLUSIONS : These findings support urgent maintenance by sending the detected distressed pavement images with the dimensions of the distressed patches and GPS coordinates to local maintenance offices in real-time.

One stage detector (RetinaNet)-based crack detection for asphalt pavements considering pavement distresses and surface objects

In this study, a supervised machine learning network model is proposed to detect and classify various types of cracks developed in asphalt pavements, including lane markers. Crack images captured from a digital camera are classified into nine categories following the pavement distress identification manual proposed by the Federal Highways Administration (FHWA). These categories are three different types of cracks, such as fatigue, longitudinal, and transverse cracks with three severity levels of the low, medium, and high for each crack type. To establish a training dataset for crack detection, 1000 images with the original size of 3704 × 10,000 pixels are divided into 20,000 smaller images of 1852 × 1000 pixels image size. The training images are labeled based on the nine categories and trained using an updated version of faster R-CNN called RetinaNet. The trained network model is validated using pavement surface images obtained from 2400 m of two road sections. It is observed from the validation study that the detection and classification accuracy of the trained network model is 84.9% considering both the crack type and severity level. When considering the crack type only, the detection accuracy of the network model is 89.1%.

Accurate detection and evaluation method for aggregate distribution uniformity of asphalt pavement

The construction uniformity of asphalt mixture has been widely evaluated, but rarely focused on the surface and internal structure of asphalt pavement simultaneously. To evaluate the aggregate distribution uniformity of asphalt pavement effectively, firstly, the pavement surface image acquisition device (PSIAD) was developed to collect the texture image on the surface of asphalt pavement. Then the sectional images of pavement core samples were acquired by industrial computed tomography (ICT) equipment. Finally, the aggregate distribution uniformity of asphalt pavement was evaluated from pavement surface and internal structure by MATLAB image processing technology. According to the results, the PSIAD based on the camera with charge coupled device (CCD) and shading black-box can avoid the information error of original pavement surface texture image, which is caused by the change of light conditions. The macro-texture width index K can substitute the texture depth to evaluate the aggregate distribution uniformity on the surface of asphalt pavement, and the aggregate distribution uniformity evaluation criterion of pavement surface based on K value was established. The improved iterative threshold segmentation algorithm based on ring-shaped partition can segment coarse aggregate particles on CT image effectively. DH is the variation coefficient of coarse aggregate area ratio in four regions on the same sectional image. DV1 is the variation coefficient of average coarse aggregate area ratio on different sectional images in one core sample. DV2 is the variation coefficient of the ratio of aggregate area with different equivalent diameter on different sectional images in one core sample. The coarse aggregate distribution evaluation indexes of DH, DV1 and DV2 can reflect the distribution uniformity of aggregate inside the asphalt pavement.

An efficient and reliable coarse-to-fine approach for asphalt pavement crack detection

Among the various defects of asphalt pavement distress, much attention has been paid to cracks which often cause significant engineering and economic problems. Crack detection is not an easy task since images of road pavement surface are very difficult to analyze. In this paper, a highly efficient pavement crack detection system is proposed, which has the following distinguishing features. Firstly, a new description of the cracks is proposed based on the spatially clustered pixels with similar gray levels. Secondly, an adaptive thresholding method is presented for image segmentation by comprehensively taking into account the spatial distribution, intensities and geometric features of cracks. Thirdly, a new concept termed Region of Belief (ROB) is introduced to facilitate the subsequent detection by defining some credibility factors which indicate the reliability that a region could be labeled as a distress region which contains cracks, and an algorithm to extract such ROBs is devised accordingly. Lastly, a novel region growing algorithm is propounded for crack detection, which features starting with an ROB seed, determining the searching scope with a specially devised rule, and searching and merging a ROB with different regions following a similarity criterion which synthetically takes different cues into consideration. Two different types of experiments were conducted. The first one was carried out using 10,000 of our field-captured images which were taken from different road conditions and environments. The second one was completed using a benchmark dataset for a comparison with other recent publications. The evaluation performance is satisfactory for a variety of different cracks. For our own data, the detection accuracy is over 95% and more than 90% of coherent cracks without disconnected fragments have been correctly detected as the integrated ones. For the benchmark data, our detection performance also outperforms previously published results. Currently, our approach has been widely applied in China.

Three-Dimensional Reconstruction of Macrotexture and Microtexture Morphology of Pavement Surface Using Six Light Sources–Based Photometric Stereo with Low-Rank Approximation

AbstractInadequate skid resistance of pavement surface is a substantial reason for traffic accidents. There is a close relationship between sliding resistance and characteristics of texture morphology, demanding high precision and comprehensive acquisition of both macrotexture and microtexture morphology. The traditional three light sources photometric stereo method is improved in this study fourfold. First, six light sources are adopted to enhance the illumination and eliminate incomplete information retrieval of pavement surface image. Second, a low-rank approximation is proposed in the image processing stage to significantly reduce the interference of noise, highlights, and shadow, resulting in a higher precision of reconstructed pavement surface compared with the existing photometric stereo method using three light sources. Third, unlike the control point–based weighting algorithm, a control point–based surface interpolation algorithm is established, which can further optimizes the precision of the re...

Multi-scale wavelet transform filtering of non-uniform pavement surface image background for automated pavement distress identification

Non-uniform background of pavement images results in difficulties when segmenting pavement images for pavement distress identification. A novel and fast non-uniform background removal algorithm based on multi-scale wavelet transform is presented. The algorithm uses multi-scale wavelet transform to decompose pavement image and then reconstructs image background using low-wavenumber components through inverse wavelet transform. Brightness of image background is then corrected to achieve uniform background pavement image. The multi-scale wavelet transform algorithm is compared with median filter algorithm and morphological closing algorithm. Experimental results show that the proposed algorithm possesses the advantage of extracting tiny cracks more effectively than the other two algorithms, demonstrating its suitability to be used in automated pavement distress segmentation and identification.

Weighted Neighborhood Pixels Segmentation Method for Automated Detection of Cracks on Pavement Surface Images

AbstractA new method is designed to detect and segment a crack on a pavement surface image from its background. Gray images of pavement surface have been collected from asphalt concrete pavement on...

Sealed-Crack Detection Algorithm Using Heuristic Thresholding Approach

AbstractSealed cracks are frequently observed on the road pavement due to pavement crack sealing. A sealed crack is a type of road surface distress, whose assessment is essential in pavement management system. In this paper, the state-of-the-art in automated pavement surface crack detection is presented. An algorithm is also developed to automatically detect sealed cracks in pavement surface images. The algorithm steps are elaborated with case study samples. Moreover, in image segmentation step, a method is developed based on a local minimum approach. The proposed algorithm is tested through various pavement surface images collected in an exhaustive case study. The evaluation is conducted with respect to the performance and robustness perspectives. The experimental results indicate that the developed algorithm has high accuracy and can consistently detect sealed cracks in different environments.

Novel Image Mosaic Algorithm for Concrete Pavement Surface Image Reconstruction

In this paper, a novel image mosaic method for concrete pavement surface image sequences reconstruction has been proposed. Harris corner points are extracted uniformly from the overlapped areas of concrete pavement surface images, which are considered feature points. The commonly used circular projection method is applied for coarse matching step and an improved point matching method is proposed for invariance of image rotation and distortion. The image fusion strategy of fading in and fading out is employed for the smooth and seamless of mosaic image. For the practical pavement surface images, which exists rotation and distortion, the corresponding experimental results show that the proposed image matching method has higher precision and stronger robustness.

Proximal Support Vector Machine Based Pavement Image Classification

Pavement cracking is one of the most important distress types. This paper provids an approach for achieving an automatic classification for pavement surface images. First, image enhancement is performed by mathematical morphological operator. secondly, pavement image segmentation is performed to separate the cracks from the background. Projection features are then extracted. The proximal support vector machine(PSVM) is used for pavement surface images classification, which is more efficient and easier to be implemented than the traditional support vector machine. The experimental results prove that the proposed method not only improves the computation efficiency but also preserves the classification performance.

An improved multifractal method for pavement cracks extraction

PurposeThis paper aims to provide an improved multifractal method to extract the pavement cracks in the complicated background. Furthermore, the pavement surface images with or without crack can also be distinguished by this method.Design/methodology/approachThe framework of analyzing the image singularity is based on the sub‐pixel multifractal measure (SPMM). Performing the SPMM can give the sub‐pixel local distribution of the image gradient and a more precise singularity exponent distribution in the image. Meantime, using the singularity exponents and the most singular manifold (MSM), the image can be decomposed into a series of sets with different statistical and physical properties automatically and easily. One can extract the cracks according to the MSM.FindingsThe example shows that the physical and geometrical properties of the pavement images can be obtained by analyzing the distribution of singularity exponents and the greatest singularity exponent. The simulation results show that the SPMM has higher quality factor in the image edge detection. And the MSM detected this way reflects the most important information of the image.Originality/valuePerforming the SPMM can give a more precise singularity exponent distribution in the image.

Wavelet-Based Pavement Distress Image Edge Detection with À Trous Algorithm

Edge detection is an alternative method in the process for identifying and classifying pavement cracks for automated pavement evaluation systems. A number of edge detectors are widely used in image processing; most specify only a spatial scale for detecting edges. However, pavement surface images frequently have various details at different scales. Therefore, wavelet-based multiscale technique can be a candidate to extract edge information from pavement surface images. Instead of detecting edges in the space domain, wavelet analysis has the ability to describe both domains in time and in frequency. It was first applied in image edge detection in 1992, using the local maximum of the magnitude of the gradient to obtain edge representation. Nevertheless, this subsampling algorithm leads to a loss of translation variance and may produce many artifacts. In this paper, wavelet edge detection based on à trous algorithm (holes algorithm) is used in pavement distress segmentation. This algorithm is an undecimated wavelet transform executed via a filter bank without subsampling process. Translation invariance is one of its most important advantages. Therefore, the algorithm can minimize the artifact in the denoised data. Results of experiments on images are discussed in the paper. By comparisons with the results derived from five other traditional edge detectors, the study demonstrates the validity and effectiveness of this method for edge detection of pavement surface distresses.

Assessment of Image-Based Data Collection and the AASHTO Provisional Standard for Cracking on Asphalt-Surfaced Pavements

The network-level pavement management system of the Kansas Department of Transportation (KDOT) is known as the Network Optimization System (NOS). For input, KDOT manually collects data on cracking severity and extent on its network annually. In this study, pavement surface images were collected and analyzed to evaluate the impact of AASHTO provisional standard PP 44-01 for asphalt-surfaced pavement-cracking data collection on NOS. The study covered approximately 262 km (164 mi) of in-service bituminous and composite pavements in Kansas. Nearly 50,000 images of the pavement surface were obtained for manual and automated evaluation of pavement condition. Two different processes of image analysis were compared with existing results from the KDOT annual survey. Transverse and fatigue crack severity and extent were manually interpreted following KDOT crack-rating algorithms for all sections. In addition, the sections were analyzed with an automated crack identification procedure following PP 44-01. The images corresponding to approximately a 5% sample rated by KDOT were identified. Comparative data were available for a manual distress survey on 5% samples (by KDOT) and image-based manual and automated interpretation on the 5% and 100% samples. Statistical analysis shows good agreement between the 5% sample and 100% sample results from the manual survey and image-based manual interpretation, respectively, for both crack types. This implies that a 5% pavement sample is adequate for describing crack severity and extent in NOS. Severity of fatigue cracks from the images was more difficult to rate, probably because of the descriptive nature of severity levels in the NOS algorithm. Agreement between the results from the manual image analysis on both 5% and 100% pavement samples and the KDOT manual survey is acceptable. However, the automated image analysis technique following the provisional AASHTO standard tends to overestimate severity of cracking.