Crack image compression sampling method based on generative adversarial network

Abstract

The invention provides a crack image compression sampling method based on a generative adversarial network. The method comprises the steps of network architecture design of a generative adversarial network, crack image generator modeling for representing a crack image and low-dimensional vector mapping relationship, adjustment and optimization of adversarial training hyper-parameters, design of acompressed observation matrix of compressed sampling, solution of an optimal low-dimensional vector and the like. According to the method, the trained crack image generator of the generative adversarial network is used as a physical constraint to realize the decompression reconstruction of the image, the sparsity of the crack image required by the traditional compressed sampling method is not required, and the application range is wider. After the generative adversarial network learns the mapping relation between the crack image and the low-dimensional vector, the low-dimensional vector is optimized based on a gradient descent method, and rapid solving of image decompression reconstruction is achieved. The method has unique advantages in the aspects of crack image reconstruction precision,reconstruction speed and the like under a relatively high compression ratio, and is relatively high in noise robustness.

Description

technical field [0001] The invention belongs to the technical field of signal processing and structural health monitoring, in particular to a crack image compression sampling method based on a generative confrontation network. Background technique [0002] At present, under the influence of long-term loads, environmental erosion and other factors, various types of infrastructure will inevitably be damaged. The continuous accumulation and development of damage will lead to the continuous decline of the bearing capacity and use function of the structure, until the safe use of the structure is endangered. Therefore, real-time monitoring of structural damage and evaluation of structural health through theoretical analysis are one of the core issues in structural health monitoring. The cracks on the surface of the structure are frequently monitored indicators, which can reflect the degree of damage to the structure and have a serious impact on the function of the structure. For ...

AI Technical Summary

Problems solved by technology

[0007] At present, the research and application of compressed sampling method based on generative adversarial network in structural health monitoring does not yet exist. It uses crack image generator to constrain decompression and reconstruction, and realizes reconstruction accuracy and noise robustness under high compression rate. The advantages of stickiness, refactoring speed, etc. have not been tapped

Images