This paper presents a road distress detection system involving the phases needed to properly deal with fully automatic road distress assessment. A vehicle equipped with line scan cameras, laser illumination and acquisition HW-SW is used to storage the digital images that will be further processed to identify road cracks. Pre-processing is firstly carried out to both smooth the texture and enhance the linear features. Non-crack features detection is then applied to mask areas of the images with joints, sealed cracks and white painting, that usually generate false positive cracking. A seed-based approach is proposed to deal with road crack detection, combining Multiple Directional Non-Minimum Suppression (MDNMS) with a symmetry check. Seeds are linked by computing the paths with the lowest cost that meet the symmetry restrictions. The whole detection process involves the use of several parameters. A correct setting becomes essential to get optimal results without manual intervention. A fully automatic approach by means of a linear SVM-based classifier ensemble able to distinguish between up to 10 different types of pavement that appear in the Spanish roads is proposed. The optimal feature vector includes different texture-based features. The parameters are then tuned depending on the output provided by the classifier. Regarding non-crack features detection, results show that the introduction of such module reduces the impact of false positives due to non-crack features up to a factor of 2. In addition, the observed performance of the crack detection system is significantly boosted by adapting the parameters to the type of pavement.

https://www.mdpi.com/1424-8220/11/10/9628/pdf

Adaptive Road Crack Detection System by Pavement Classification

Pavement condition information is a significant component in Pavement Management Systems. The labeling and quantification of the type, severity, and extent of surface cracking is a challenging area for weighing the asphalt pavements. This paper presents a widespread review on various platform and image processing approaches for asphalt surface interpretation. The main part of this study presents a comprehensive combination of the state of the art in image processing based on crack interpretation related to asphalt pavements. An attempt is made to study the existing methodologies from different points of views accompanied by extensive comparisons on three stages of methods—distress detection, classification, and quantification to facilitate further research studies. This paper presents a survey of the developed pavement inspection systems up to date. Additionally, emerging and evolution technologies considered to automate the processes are discussed.

Image Based Techniques for Crack Detection, Classification and Quantification in Asphalt Pavement: A Review

Investigating Driver Injury Severity Patterns in Rollover Crashes Using Support Vector Machine Models

Summary

Visual inspection procedures remain the primary method of infrastructure assessment throughout the USA, but their shortcomings are numerous. In addition to their widely acknowledged variability and subjectivity, the large scale of civil infrastructure systems presents expensive access and time requirements that constrain the frequency of visual inspections and result in poor temporal resolution, which hampers effective decision-making. To overcome this challenge, the research reported herein aimed to assess the ability of computer algorithms together with imagery collected by unmanned aerial vehicles (UAV) to extract accurate and quantitative information to help inform infrastructure management decisions. Techniques such as homography and lens distortion correction are used in this article in a post-processing framework that allows the use of color images obtained by UAVs for actual damage quantification measurements. The experiments described in this article utilize a UAV with a mounted camera and provide measurements from a representative infrastructure mockup with several simulated damage scenarios. Deformation measurements, change detection (related to structural features and the size of deterioration), and crack pattern identification were all analyzed. The results indicated that the developed post-processing algorithms were able to extract quantitative information from UAV captured imagery. Copyright © 2016 John Wiley & Sons, Ltd.

Bridge related damage quantification using unmanned aerial vehicle imagery

Kernel methods are algorithms that, by replacing the inner product with an appropriate positive definite function, implicitly perform a nonlinear mapping of the input data into a high-dimensional feature space. In this paper, we present a kernel method for clustering inspired by the classical k-means algorithm in which each cluster is iteratively refined using a one-class support vector machine. Our method, which can be easily implemented, compares favorably with respect to popular clustering algorithms, like k-means, neural gas, and self-organizing maps, on a synthetic data set and three UCI real data benchmarks (IRIS data, Wisconsin breast cancer database, Spam database).

A Novel Kernel Method for Clustering.

Effect of lifters and mill speed on particle behaviour, torque, and power consumption of a tumbling ball mill: Experimental study and DEM simulation

Global optimization of reliability design for large ball mill gear transmission based on the Kriging model and genetic algorithm

Reliability design and optimization of the planetary gear by a GA based on the DEM and Kriging model

This paper describes a dynamic model with four degrees of freedom for simulating the rollover stability performances of articulated loaders. The dynamic model considers three translational degrees of freedom and a rotational degree of freedom about the tilting axis and was verified by field experiments and virtual prototype experiments. The field experiments using a scaled articulated loader are detailed, and the experimental data of the lateral load transfer ratio, the lateral acceleration and the roll angle of the vehicle are provided for classification at different speeds and different turning radii. The virtual prototype experiments are modelled on the experimental model as well as the simulation conditions. Comparing the experimental data with the theoretical data shows that the dynamics model is reasonable and that the virtual prototype simulation can be used instead of the field experiments to improve the security of active safety technology development.

Dynamic simulation for the rollover stability performances of articulated vehicles

This paper devotes both analytical and experimental efforts in developing a comprehensive dynamic model for an articulated steering wheel loader. The general motion of a wheel loader without suspension is described by seven degrees of freedom (DOF) (three for translation and four for rotation) in this model, which can be used to study the problem of wheel loader dynamics on slopes and over obstacles. A scale wheel loader was designed and manufactured to validate the dynamic model in three conditions (turning on level ground, turning on slopes, and passing over obstacles). The test results reasonably agree with the simulation results. The developed dynamic model was found to be useful and could serve as an important tool for analysing the stability of wheel loaders.

Guoqiang Wang

Papers

Effect of lifters and mill speed on particle behaviour, torque, and power consumption of a tumbling ball mill: Experimental study and DEM simulation

Global optimization of reliability design for large ball mill gear transmission based on the Kriging model and genetic algorithm

Reliability design and optimization of the planetary gear by a GA based on the DEM and Kriging model

Dynamic simulation for the rollover stability performances of articulated vehicles

Dynamic model and validation of an articulated steering wheel loader on slopes and over obstacles