Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A Fast Parallel Implementation of Nonlocal Mean Filtering

A technology of non-local mean and realization method, applied in image data processing, instrumentation, calculation, etc.

Active Publication Date: 2016-05-25
江苏一影医疗设备有限公司
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The final output image is In summary, the computational complexity of the common GPU acceleration algorithm is O(|N|((2B+1)(2B+1)+2)+1)

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
  • A Fast Parallel Implementation of Nonlocal Mean Filtering
  • A Fast Parallel Implementation of Nonlocal Mean Filtering
  • A Fast Parallel Implementation of Nonlocal Mean Filtering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] In the following, the present invention will be further clarified in conjunction with specific examples. It should be understood that these examples are only used to illustrate the present invention and not to limit the scope of the present invention. After reading the present invention, those skilled in the art will give various equivalent forms of the present invention. All the modifications fall within the scope defined by the appended claims of this application.

[0043] The fast non-local mean filtering algorithm based on improved GPU parallelism includes the following steps:

[0044] Step 1. In the GPU, each thread calculates the absolute value of the grayscale difference between its corresponding pixel and the pixel at a certain position in the search window. After all the threads have calculated the difference value, calculate the B+1 possible grayscale cumulative value of the center row of the comparison block (assuming radius B) centered on the pixel corresponding ...

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

A non-local average filtering algorithm is an algorithm which is extensively used for pattern noise restraining. The algorithm constructs a weighting filter to restrain noises in images according to the hypothesis that similar neighborhood structures exist around pixels belonging to a same image structure and on the basis of neighborhood similarity. Experimental results show that the non-local average filtering algorithm can effectively restrain the noises in the images while keeping organization information of the images; in order to effectively restrain noises in the images, generally, a larger search window is required to lead into more neighborhood information, amount of computing work and processing time are required, and the application in the reality is influenced. For solving the problems, the invention provides a fast parallel achieving method for non-local average filtering. According to the invention, shared storage properties and non-local average weight symmetry are used for optimizing parallel operation on the basis of original GPU parallel using pixel as a unit, and the computation speed of the non-local average filtering algorithm is increased remarkably.

Description

Technical field [0001] The invention relates to a fast parallel realization method of a non-local mean filtering algorithm on a GPU. Background technique [0002] Image noise reduction has always been an important research content in the field of digital image processing. The classic noise reduction filtering methods include neighborhood average method, median method and some frequency domain filtering methods. These image noise reduction algorithms are generally based on pixel gray Information such as degree difference and gradient only uses information in a small neighborhood, which is likely to lead to blurred image processing results. Buades proposed a non-local mean filtering algorithm based on the fact that any small window from the image can be found in a larger area of ​​the image and many similar window structures can be found. This algorithm can make full use of the changes in the image. The image information in a large range suppresses noise, so that the noise in the ...

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 Patents(China)
IPC IPC(8): G06T5/00
Inventor 陈阳庄志昆罗立民李松毅鲍旭东
Owner 江苏一影医疗设备有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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