Search

Crack is one of the important pavement damages. The number and types of cracks are usually collected and inspected by human, but lately computer vision has shown great potential for automatically identifying pavement cracks. However, training a convolutional neural network (CNN) requires large amount of data. The limited number of crack images and the variety of crack image background put challenges for developing CNN model with high accuracy. This study proposes an intelligent method for crack pattern classification based on limited field images. The proposed method has a two-step data preprocessing. It first applied data augmentation method to enlarge the dataset, solving the data imbalance problem and providing more training data. Then a crack extraction method was applied to convert the original image into a binary back-and-white image. This step significantly reduced the input feature and simplified the model. Both of data preprocessing steps were designed to decrease the bias of the model. Then we performed a model selection and hyper-parameter tunning for CNN. We explored the application of AlexNet, SE-Net, and ResNet with a variety of configurations. The results show that the proposed data augmentation can enlarge the dataset significantly. The crack extraction also significantly increased the test accuracy. The ResNet with 50 layers has the highest test accuracy. With data augmentation, crack extraction and model selection combining together, the final approach can improve the test accuracy from 52.68% to 87.50% compared to the original limited images. The result suggests our proposed method works well to classify crack images of asphalt pavement with limited sample size. [ABSTRACT FROM AUTHOR]

In this work, the study of a complex masonry umbrella vault located within Masegra Castle (Museum of Sondrio territory, Northern Italy) is presented. Built during the renewal of the castle in the 16th century, the umbrella vault of Masegra Castle is characterized by a total of eight sails, with four half-cross vaults added at the corners, supported by four walls, and resulting in a quadrangular horizontal plan. The observed crack pattern seems to derive from vertical settlement phenomena involving at least one of the supporting walls. A multidisciplinary approach based on innovative technologies for the geometric reconstruction, detailed analysis of the construction, the historical evolution, and its state of conservation, has been followed. Starting from dense laser scanner point clouds and photogrammetric surveys, historical building information modeling (HBIM) models have been created and allowed the generation of an accurate geometrical representation of the umbrella vault through nonuniform rational basis-splines (NURBS) surfaces. Starting from the NURBS-based model, an adaptive kinematic procedure is used to determine the failure behavior under settlements. A discretization through triangular NURBS elements, in which the real geometry remains unvaried, is adopted. Each element is idealized as a rigid block and jumps of displacement are allowed at the boundaries only. By applying the principle of virtual works, a discontinuous displacement field deriving from an imposed settlement is obtained. As a consequence of the reduced number of elements adopted, a metaheuristic mesh adaptation procedure is finally applied to find the correct disposition of fracture lines. Different settlement typologies are adopted in the analyses to reproduce the crack pattern observed on the umbrella vault. Finally, an inverse analysis is conducted to estimate the shape of the occurred settlements starting from the observed crack pattern. [ABSTRACT FROM AUTHOR]