Deep convolutional generative adversarial network-based synthesis of datasets for road pavement distress segmentation

We discuss a range of problems relating to road pavement defects detection and modern approaches their solution. The presented comparison publicly available datasets allows one make conclusion that the problem segmentation in driver wide-view images is difficult poorly investigated. To solve this problem, we have developed algorithms for generating synthetic dataset cracks potholes distress based on computer graphics methods deep convolutional generative adversarial networks. A accuracy was performed by training fully neural network U-Net real combined datasets.