Generative adversarial network-based bridge crack image generation model

Abstract

The invention relates to a generative adversarial network-based bridge crack image generation model. The bridge crack image generation model comprises a judgment sub-model and a generation sub-model;the judgment sub-model adopts six convolutional layers; each of the first to fifth convolutional layers adopts a 5*5 convolutional kernel; the sixth convolutional layer adopts a 1*1 convolutional kernel; the generation sub-model of the bridge crack generation model comprises five deconvolutional layers; and each convolutional layer adopts the 5*5 convolutional kernel. Multiple convolutional kernels are used, so that during convolutional operation, the situation of lack of continuity of integral information of a picture due to the excessively small convolutional kernel and the problem of loss of a large amount of picture details due to the excessively large convolutional kernel are avoided; and features learnt by channels are fused, so that the problem of no information relationship among feature graphs can be improved, and meanwhile, the restoration effect of image subsequent processing is improved.

Description

technical field [0001] The invention belongs to the technical field of computer vision, in particular to a bridge crack image generation model based on a generative confrontation network. Background technique [0002] With the development of the transportation industry, road maintenance has become very important. As an important part of today's social transportation hub, bridges not only undertake the heavy responsibility of transportation, but also affect the safety of transportation personnel. However, due to the bridge structure's inevitable long-term use, various damages will occur, resulting in bridge structure resistance attenuation and safety hazards. Therefore, regular inspection and maintenance are essential. Cracks are the most common defect in bridges. Bridge cracks can occur for a variety of reasons, mainly due to fatigue of the asphalt pavement, combined with unfavorable atmospheric conditions that may cause material shrinkage, or due to poor quality asphalt m...

AI Technical Summary

Problems solved by technology

[0007] In order to solve the above-mentioned problems existing in the prior art, the present invention provides a bridge crack image generation model based on a generative confrontation network, which uses multiple convolution kernels to avoid the problem of too small convolution kernels when performing convolution operations. The lack of continuity of the overall information and the loss of a large number of image details caused by too large a convolution kernel, the fusion of the features learned by each channel can improve the problem of information disconnection between each feature map, and at the same time, improve the image quality. Restoration effect of subsequent processing

Images