GPU-based high resolution image real-time enhancement method

A high-resolution image and image technology, applied in the field of image processing, can solve problems such as inability to meet application requirements, difficult to ensure real-time performance, and large amount of calculation, to optimize register usage and algorithm flow, solve adjacent pixel interference, and fast. The effect of removing noise

Active Publication Date: 2019-02-15
XIDIAN UNIV
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the algorithm involves a lot of filtering operations, and the amount of calculation is very large. With the increase of image resolution and the number of reconstruction iterations, the traditional CPU-based implementation method cannot guarantee real-time performance and cannot meet the actual application requirements.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • GPU-based high resolution image real-time enhancement method
  • GPU-based high resolution image real-time enhancement method
  • GPU-based high resolution image real-time enhancement method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0057]In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0058] Embodiments of the present invention provide a GPU-based real-time enhancement method for high-resolution images, such as figure 1 As shown, the method is implemented through the following steps:

[0059] Step 1: Read an image to be enhanced to the host memory;

[0060] Specifically, the resolutions of images to be enhanced are 512×512, 1024×512, 1024×1024, 2048×1024, 2048×2048, 4096×2048, and 4096×4096, and 0%-5% salt and pepper noise is added to the image to be enhanced , to verify the accuracy of the parallel algorithm. Depend on figure 2 It can be seen that the image to be enhanced i...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a GPU-based high resolution image real-time enhancement method, comprising the steps of performing the Gaussian curvature filtering on the input image, then reconstructing theinput image according to the gradient of the filtered input image combined with the parallel Gaussian curvature filtering based on coordinate separation, and then subjecting the reconstructed input image to noise suppression to obtain an enhanced image. The parallel Gaussian curvature filter parallel implementation method based on coordinate separation solves the problem of adjacent pixels interference, accelerates the filter convergence speed, and can remove the noise faster than the traditional method of synchronously updating all pixels under the same iteration number, and improves the image quality.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a GPU-based real-time high-resolution image enhancement method. Background technique [0002] In recent years, with the rapid development of semiconductor technology, the computing power of graphics processing unit (GPU) has been greatly improved. Compared with CPU, it has stronger arithmetic computing power and higher memory bandwidth. Parallel processing has great advantages; at the same time, with the release of CUDA and OpenCL and other development tools, the programming difficulty of parallel computing has been further reduced, which makes GPU-based parallel computing receive extensive attention in the fields of machine learning and computer vision. It has gradually become a research hotspot in the field of image processing. [0003] The image enhancement algorithm based on Gaussian curvature filtering and gradient field reconstruction enhances image co...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06T5/00G06T1/20
CPCG06T1/20G06T5/002G06T2207/20024G06T2200/28G06T2207/10004Y02D10/00
Inventor 周慧鑫黄楙森成宽洪赵东宋江鲁奇于跃李欢姚博秦翰林杜娟宋尚真谭威
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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