A Holistic Approach for Inspection of Civil Infrastructures Based on Computer Vision Techniques

Abstract

ABSTRACT:In this work, it is examined the 2D recognition and 3D modelling of concrete tunnel cracks, through visual cues. At the time being, the structural integrity inspection of large-scale infrastructures is mainly performed through visual observations by human inspectors, who identify structural defects, rate them and, then, categorize their severity. The described approach targets at minimum human intervention, for autonomous inspection of civil infrastructures. The shortfalls of existing approaches in crack assessment are being addressed by proposing a novel detection scheme. Although efforts have been made in the field, synergies among proposed techniques are still missing. The holistic approach of this paper exploits the state of the art techniques of pattern recognition and stereo-matching, in order to build accurate 3D crack models. The innovation lies in the hybrid approach for the CNN detector initialization, and the use of the modified census transformation for stereo matching along with a binary fusion of two state-of-the-art optimization schemes. The described approach manages to deal with images of harsh radiometry, along with severe radiometric differences in the stereo pair. The effectiveness of this workflow is evaluated on a real dataset gathered in highway and railway tunnels. What is promising is that the computer vision workflow described in this work can be transferred, with adaptations of course, to other infrastructure such as pipelines, bridges and large industrial facilities that are in the need of continuous state assessment during their operational life cycle.

Modern Crack Detection for Bridge Infrastructure Maintenance Using Machine Learning

“…Using an insufficient set of samples for training can lead to overfitting [29]. Additionally, the collection of abundant samples is an exorbitantly costly endeavour, which has increased the utility of data augmentation methods (i.e., flipping, resizing, random cropping) to enhance image variation and overcome the issue of over-fitting [34]. In the overall training procedure of the proposed approach, label generation and crack detection were performed through data augmentation are presented in the Table 3.…”

“…In recent years, Generative adversarial networks (GAN) have been employed in deep learning methods successfully as it offers a novel strategy for the training of different models [34]. Originally for GAN, a fully connected layered generator configuration is used that allows the images to be generated from random noises.…”

Deep-learning-based crack detection with applications for the structural health monitoring of gas turbines

“…Stentoumis et al 5 performed a detailed study on ''A Holistic Approach for Inspection of Civil Infrastructures Based on Computer Vision Techniques.'' Stentoumis et al research was conducted on concrete tunnel cracks using computer vision techniques and CNNs.…”

“…Stentoumis et al 5 performed a detailed study on “A Holistic Approach for Inspection of Civil Infrastructures Based on Computer Vision Techniques.” Stentoumis et al research was conducted on concrete tunnel cracks using computer vision techniques and CNNs. In this work, several IPTs were used as pre-processing steps for CNNs, including noise removal, straight lines removal, curve detection, shape filtering through a Hough transform, and morphological reconstruction.…”

“…CNNs are known to perform much better than traditional computer vision methods and are very useful in applications such as image classification, object detection, and segmentation. 4,5 With the development of CNNs, the detection of surface cracks can be performed without further image processing like Cha et al, 6 Krizhevsky et al, 7 and Zhang et al 8 due to the fact that they are capable of learning image features automatically. Deep learning models exploit image features in different resolutions through stacked convolutional layers, which leads to an improved detection and classification performance.…”