Image quality evaluation method and device, computer equipment, readable storage medium and program product

By performing feature analysis and multisemantic feature generation of the evaluated image, combined with image gradient, brightness and contrast characteristics, the problem of the existing reference-free image quality evaluation algorithm poor effect in real distorted images is solved, and more accurate image quality judgment is achieved.

CN119648664APending Publication Date: 2025-03-18CHINA TELECOM CLOUD TECH CO LTD
0 Cites 1 Cited by

Patent Information

Application Number
CN202411748809.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-12-02
Publication Date
2025-03-18

AI Technical Summary

Technical Problem

The existing reference-free image quality evaluation algorithm is poor when facing real distorted images, which affects the accuracy of image quality judgment.

Method used

By performing feature analysis on the image to be evaluated, image features, image gradient features, image brightness features and image contrast features are obtained, and multisemantic features are generated based on these features. Multisemantic fusion features are generated through feature fusion, and image quality is finally determined based on multisemantic fusion features.

Benefits of technology

It improves the accuracy of image quality evaluation, can supplement original image information from different dimensions of image gradient, brightness and contrast, and enhances the ability to judge image quality.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to an image quality evaluation method and device, computer equipment, a readable storage medium and a program product. The method comprises the following steps: performing feature analysis on a to-be-evaluated image to obtain an image feature, an image gradient feature, an image brightness feature and an image contrast feature of the to-be-evaluated image, performing feature fusion on the features to obtain a multi-semantic fusion feature, and determining the image quality of the to-be-evaluated image according to the multi-semantic fusion feature. The multi-semantic fusion feature is a mixed feature obtained by performing feature fusion on the image feature of the to-be-evaluated image and the multi-semantic feature, and the multi-semantic feature is generated according to the image gradient feature, the image brightness feature and the image contrast feature of the to-be-evaluated image, so that the multi-semantic fusion feature is more accurate. The multi-semantic fusion feature fused with the multi-semantic feature can supplement the original image information from different dimensions of image gradient, image brightness and image contrast, so that the accuracy of determining the image quality based on the multi-semantic fusion feature can be improved.
Need to check novelty before this filing date? Find Prior Art