DeepCrack: Learning Hierarchical Convolutional Features for Crack Detection

Short summary

DeepCrack, a novel deep convolutional neural network, achieves an average F-Measure over 0.87 for automatic crack detection by fusing multi-scale hierarchical features, outperforming state-of-the-art methods on challenging datasets.

AI-generated from the title and abstract; the full text is not read.