Super-resolution reconstruction method and device based on soil CT image

Abstract

The invention provides a super-resolution reconstruction method and device based on a soil CT image. The method comprises the following steps: acquiring an original high-resolution CT image; performing down-sampling processing on the original high-resolution CT image to obtain a low-resolution CT sequence image; and inputting the low-resolution CT sequence image into a sequence image generator model to obtain a super-resolution reconstructed image output by the sequence image generator model, wherein the sequence image generator model is obtained by performing adversarial training based on a generative adversarial network formed by the sequence image generator model and a discriminator model, and the discriminator model is used for discriminating the image credibility. The deep learning neural network-based sequence image generator model is used for performing super-resolution reconstruction on the soil CT image, so that the problems of low definition of the soil CT image and fuzzy pore boundary are solved, and clear and reliable data are provided for soil morphology research.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a soil CT image-based super-resolution reconstruction method and device. Background technique [0002] The super-resolution reconstruction methods currently applied to CT images mainly include the three-dimensional convex set projection method, the three-dimensional high-frequency correction A+ method, and so on. In the prior art, soil CT images have problems such as low definition and blurred pore boundaries, and there is no research on applying super-resolution reconstruction methods to soil CT images. [0003] Traditional super-resolution methods have problems such as low definition of super-resolution reconstructed images and the need to manually determine model parameters. Among them, the 3D convex set projection method needs to manually determine the correction threshold, which cannot realize image adaptive super-resolution reconstruction; the 3D high-freq...

AI Technical Summary

Problems solved by technology

[0005] The invention provides a super-resolution reconstruction method and device based on soil CT images, which are used to solve the defects of low definition of soil CT images and blurred pore boundaries in the prior art, and realize deep learning-based reconstruction that can combine image sequence information. Soil CT image super-resolution method

Images