Automatic Road Defect Detection by Textural Pattern Recognition Based on AdaBoost

Abstract

The state of roads is continuously degrading due to meteorological conditions, ground movements, and traffic, leading to the formation of defects, such as grabbing, holes, and cracks. In this article, a method to automatically distinguish images of road surfaces with defects from road surfaces without defects is presented. This method, based on supervised learning, is generic and may be applied to all type of defects present in those images. They typically present strong textural information with patterns that show fluctuations at small scales and some uniformity at larger scales. The textural information is described by applying a large set of linear and nonlinear filters. To select the most pertinent ones for the current application, a supervised learning based on AdaBoost is performed. The whole process is tested both on a textural recognition task based on the VisTex image database and on road images collected by a dedicated road imaging system. A comparison with a recent cracks detection algorithm from Oliveira and Correia demonstrates the proposed method's efficiency

A Fast Detection Method via Region‐Based Fully Convolutional Neural Networks for Shield Tunnel Lining Defects

“…In general, the implementation of these methods involves the following three steps: (1) contrast enhancement, (2) mathematical morphological processing, and (3) information extraction using linear filters. Other machine learning methods, such as artificial neural networks (ANN) (Adeli and Yeh, ; Eldin and Senouci, ; Jin and Zhou, ), support vector machine (SVM) (Qu et al., ), Adaboost (Cord and Chambon, ), K‐nearest neighbors algorithm (Lei and Zuo, ), grouping techniques (Yeum and Dyke, ) and Restricted Boltzmann Machine (Rafiei and Adeli, , ; Rafiei et al., ) have also been used in the field of civil engineering for crack or damage detection and achieved some good results. However, a common problem with these methods is the inability to handle complex background images.…”

Edge‐Enhanced Matching for Gradient‐Based Computer Vision Displacement Measurement

“…Vision sensors, among these new techniques, have been broadly applied for civil engineering problems. Famous applications of vision‐sensing techniques include dynamic displacement monitoring (Cha et al., ; Park et al., ; Yoon et al., ), three‐axes (i.e., X‐axis, Y‐axis, and depth) displacement measurement (Park et al., ; Abdelbarr et al., ), surface displacement/strain measurement (Luo et al., ; Almeida et al., ), vision‐based structural analysis (Chen et al., ; Sharif et al., ; Park et al., ), cable tensile force evaluation (Kim et al., ), bridge‐lining inspection (Zhu et al., ), rocking motion and landslide monitoring (Debella‐Gilo and Kääb, ; Greenbaum et al., ), automatic construction progress assessment (Bügler et al., ), 3D object finding in point cloud (Sharif et al., ), surface crack/defection detection based on texture‐based video processing (Cord and Chambon, ; Chen et al., ) or deep learning (Cha et al., ; Cha et al, ; Zhang et al., ), vehicle classification based on spectrogram features (Yeum et al., ), and intelligent transportation (Chen et al., ; Fernandez‐Llorca et al., ). With advancement in image sensors and computer techniques such as computer vision, cloud computing, and wireless data transfer, vision sensors have become more cost‐effective and computation‐efficient, thus have high potential in field application for SHM problems.…”

An ensemble machine learning‐based modeling framework for analysis of traffic crash frequency

“…It adjusts weights to each observation and each base learner using iterative training, reducing both the variance and the bias (Cord & Chambon, 2012). In contrast to averaging methods, AdaBoosting provides sequential learning of predictors.…”