Unlock instant, AI-driven research and patent intelligence for your innovation.

Nonlocal Uniform Digital Image Denoising Method Based on Mahalanobis Distance

A Mahalanobis distance, digital image technology, applied in image enhancement, image data processing, instruments, etc., can solve the problem of not considering the difference in pixel value distribution of image blocks, affecting the image denoising effect of non-local mean algorithm, and cannot exclude data correlation. It can solve problems such as sexual interference, etc., to achieve the effect of easy viewing, improved signal-to-noise ratio, and enhanced resolution.

Inactive Publication Date: 2019-06-25
NANJING UNIV OF SCI & TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

), the algorithm has disadvantages: it has a large amount of calculation; it uses Euclidean distance to measure the similarity of image blocks, and does not consider the difference in the distribution of pixel values ​​​​of image blocks, nor can it rule out the interference of data correlation
These defects directly affect the actual image denoising effect of the non-local mean algorithm

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
  • Nonlocal Uniform Digital Image Denoising Method Based on Mahalanobis Distance
  • Nonlocal Uniform Digital Image Denoising Method Based on Mahalanobis Distance
  • Nonlocal Uniform Digital Image Denoising Method Based on Mahalanobis Distance

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0049] The present invention is described below with embodiment,

[0050] to combine figure 1 , the present invention is based on the Mahalanobis distance non-local means digital image denoising method, and the steps are as follows:

[0051] Step 1, follow figure 2The coordinates of a certain pixel point in the two-dimensional digital image, select a window of L×L (L takes 11) around it.

[0052] Step 2, form a column vector x for 5×5 neighboring points of each pixel in the 11×11 window, and the elements in the vector are the pixel values ​​of the pixel.

[0053] Step 3, for the vector x of the adjacent points of each pixel in the window j Neighboring point vector x with the center pixel i Calculate the covariance matrix S, the calculation formula is:

[0054] S=(x ij -x 0 )(x ij -x 0 ) T

[0055] where x ij =(x i x j ), T Represents the transpose operation on the matrix;

[0056] Step 4: Perform singular value decomposition on S:

[0057]

[0058] wher...

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 nonlocal uniform digital image denoising method based on Mahalanobis distance. On a basis of a nonlocal mean method, a method of measuring image block similarity by using the Mahalanobis distance to replace Euclidean distance in a window neighborhood is provided, and by considering the instability of the Mahalanobis distance, a Moore-Penrose inverse matrix theory is used to improve the Mahalanobis distance, and then a stable Mahalanobis distance calculation method is acquired; a Gaussian kernel adopting the Mahalanobis distance as a filter coefficient is used for weighted mean of pixel values in an image, and then the signal-to-noise ratio of the image is improved. The nonlocal uniform digital image denoising method is advantageous in that an anti-noise performance of an algorithm is enhanced, the signal-to-noise ratio of the image is improved obviously, and at the same time, the detail information of the image is kept, and the resolution of the image is enhanced.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, in particular to a non-local uniform digital image denoising method based on Mahalanobis distance. Background technique [0002] With the development of optoelectronic technology and the widespread use of various electronic display devices, people have higher and higher requirements for the imaging quality of digital images. However, image information will be disturbed by a lot of noise during the process of acquisition, transmission and recording. Therefore, before the digital image is finally displayed, the noise in the digital image must be removed first. [0003] The previous digital image denoising technology will cause the loss of detail information in the image while removing the noise, and the denoising effect is not ideal. For example, common Gaussian filtering and bilateral filtering will blur the filtered image and reduce the image resolution. The non-local uniform ...

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 路东明阴盼强顾国华钱惟贤任侃陈钱赵蓉丁祺牛世伟孔维一刘晗霜
Owner NANJING UNIV OF SCI & TECH