No-reference image quality evaluation method based on convolutional neural network
A technology of convolutional neural network and reference image, which is applied in the application field of generative confrontation network in image quality evaluation, can solve the problems of low performance and poor performance of no reference evaluation method, and achieve the effect of fast training speed and good experimental effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0019]The present invention will be further described below.
[0020]A non-reference image quality evaluation method based on convolutional neural network, the specific implementation steps are as follows:
[0021]Step 1: Preprocessing the distortion map and the natural map to get a similar image;
[0022]1-1. Calculate the brightness comparison:
[0023]For distortion diagrams: x, natural map: y, adoptionIndicates the brightness information of the distortion diagram x,Indicates the brightness information of the natural map Y:
[0024]
[0025]Where XiYiThe pixel point value of the image.
[0026]Then the brightness comparison of the distortion diagram X and the natural map Y can be represented as:
[0027]
[0028]Where C1It is to prevent the minimum number of points set by 0.
[0029]1-2. Comparative contrast contrast: c (x, y)
[0030]ΣxContrast information indicating distortion diagram x, σyRepresents contrast information of natural map Y:
[0031]
[0032]The contrast of the distortion diagram X and the natural map ...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap