Dam crack detection based on deep learning cascade detection algorithm

In order to improve the crack detection accuracy of concrete dam surface, a cascade crack detection algorithm based on deep learning and prior information reinforcement is proposed. In the algorithm, two neural networks are cascaded to detect cracks by a cascade detection method. Prior information strengthening includes standardization and edge strengthening. When making standardized crack slices, the crack is rotated to the horizontal trend and the center of mass is placed in the section center. Meanwhile, slices belonging to the general background are removed according to the gray gradient similarity. The slices were processed with the prior information of crack edges to artificially highlight the features and reduce the difficulty of network training. The network results without prior information reinforcement and cascade detection are compared respectively to evaluate the network detection performance. The experimental results show that the accuracy rate and recall rate of concrete dam surface crack detection have been improved significantly, which proves the effectiveness of the algorithm.