Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 2 of 2
Full-Text Articles in Engineering
Comparison Of Deep Convolutional Neural Networks And Edge Detectors For Image-Based Crack Detection In Concrete, Sattar Dorafshan, Robert J. Thomas, Marc Maguire
Comparison Of Deep Convolutional Neural Networks And Edge Detectors For Image-Based Crack Detection In Concrete, Sattar Dorafshan, Robert J. Thomas, Marc Maguire
Civil and Environmental Engineering Faculty Publications
This paper compares the performance of common edge detectors and deep convolutional neural networks (DCNN) for image-based crack detection in concrete structures. A dataset of 19 high definition images (3420 sub-images, 319 with cracks and 3101 without) of concrete is analyzed using six common edge detection schemes (Roberts, Prewitt, Sobel, Laplacian of Gaussian, Butterworth, and Gaussian) and using the AlexNet DCNN architecture in fully trained, transfer learning, and classifier modes. The relative performance of each crack detection method is compared here for the first time on a single dataset. Edge detection methods accurately detected 53–79% of cracked pixels, but they …
Charpy Impact Test Methods For Cementitious Composites: Review And Commentary, Robert J. Thomas, Andrew D. Sorensen
Charpy Impact Test Methods For Cementitious Composites: Review And Commentary, Robert J. Thomas, Andrew D. Sorensen
Civil and Environmental Engineering Faculty Publications
Several researchers have recently employed the Charpy method to characterize the high strain rate mechanical strength of cementitious composites. This paper provides a critical review of existing applications of the Charpy method for impact testing of cementitious composites. Studies have employed various specimen sizes and geometries. Additionally, some studies have tested notched specimens while others have tested plain. Furthermore, varying methods of normalization result in results reported in a variety of incompatible units. The lack of consistency between studies limits the basis for comparison and the ability to validate results, which demonstrates a clear need for a standardized method for …