Image super-resolution method based on deep threshold convolutional neural network
A convolutional neural network and super-resolution technology, applied in graphics and image conversion, image data processing, instruments, etc., can solve the problems of model learning effect decline, short time, etc., achieve fast speed, reduce gradient disappearance, and fast high-resolution images Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0029] The embodiments and effects of the present invention will be described in further detail below in conjunction with the accompanying drawings.
[0030] refer to figure 1 , the realization steps of the present invention are as follows.
[0031] Step 1: Obtain pairs of low-resolution and high-resolution image data.
[0032] 1.1) Obtain low-resolution images:
[0033] The original image is first down-sampled, and then the down-sampled image is restored to its original size by bilinear cubic interpolation, and the obtained picture is a low-resolution image;
[0034] The bilinear cubic interpolation is performed by the following formula:
[0035] f(i+u,j+v)=ABC
[0036] Among them, u represents the horizontal interpolation position, v represents the vertical interpolation position, i is the abscissa of the current pixel, j is the ordinate of the current pixel, f(i+u,j+v) indicates that the image is in (i+ The interpolated pixel value at u,j+v); A is the horizontal factor...
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, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com