Method, apparatus, terminal and server for evaluating image quality

Abstract

The invention provides an image quality evaluation method, a device, a terminal and a server. The method includes obtaining a suspected standard image and a reference image set. Suspected standard images are used as evaluation criteria to obtain the quality score of each reference image in the reference image set, and the corresponding relationship between the reference image and the quality scoreis obtained. A training set is constructed according to the reference image and the quality score, a training quality evaluation model is established, and the quality of the image to be evaluated isevaluated according to the quality evaluation model. By training the quality evaluation model, the invention can give an objective score to any image on the premise that the standard image is not provided, so as to accurately judge whether the image meets the specific requirements of industry or security.

Description

technical field [0001] The invention relates to the field of computers, in particular to an image quality evaluation method, device, terminal and server. Background technique [0002] In the field of image evaluation technology, there are mainly evaluation methods based on PSNR and evaluation methods based on SSIM in the prior art. However, these two evaluation methods cannot reflect the real actual image quality, are difficult to apply to complex environments, and are difficult to meet security requirements. In addition, in the prior art, image quality evaluation relies on standard images, and standard images have high requirements, so they cannot be obtained in many application scenarios, thus limiting the application range of image quality evaluation methods. Contents of the invention [0003] In order to solve the above technical problems, the present invention proposes an image quality evaluation method, device, terminal and server. The present invention is specifica...

AI Technical Summary

Problems solved by technology

[0002] In the field of image evaluation technology, there are mainly PSNR-based evaluation methods and SSIM-based evaluation methods in the prior art, but these two evaluation methods cannot reflect the real actual image quality, are difficult to apply to complex environments, and are difficult to meet security requirements.

In addition, in the prior art, image quality evaluation relies on standard images, and standard images have high requirements, so they cannot be obtained in many application scenarios, thus limiting the application range of image quality evaluation methods

Images