CN108320269A - A kind of efficient method for eliminating high density salt-pepper noise - Google Patents
A kind of efficient method for eliminating high density salt-pepper noise Download PDFInfo
- Publication number
- CN108320269A CN108320269A CN201710041140.7A CN201710041140A CN108320269A CN 108320269 A CN108320269 A CN 108320269A CN 201710041140 A CN201710041140 A CN 201710041140A CN 108320269 A CN108320269 A CN 108320269A
- Authority
- CN
- China
- Prior art keywords
- salt
- median
- pixel
- pepper noise
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000006002 Pepper Substances 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000001914 filtration Methods 0.000 claims abstract description 20
- 150000003839 salts Chemical class 0.000 claims description 25
- 230000001419 dependent effect Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 5
- 230000003044 adaptive effect Effects 0.000 description 4
- 235000002566 Capsicum Nutrition 0.000 description 3
- 241000722363 Piper Species 0.000 description 3
- 235000016761 Piper aduncum Nutrition 0.000 description 3
- 235000017804 Piper guineense Nutrition 0.000 description 3
- 235000008184 Piper nigrum Nutrition 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000010850 salt effect Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of efficient methods for eliminating high density salt-pepper noise, include the following steps:1) salt-pepper noise in image is demarcated;2) intermediate value filtering, and the method that the calculating of median is divided using mean value are carried out to the salt-pepper noise of calibration.The present invention only filters out the salt-pepper noise of calibration, rather than is filtered to all pixels, therefore the filtering time of the invention is solely dependent upon the density of salt-pepper noise;Secondly intermediate value filtering is carried out to salt-pepper noise, the edge details of image can be preferably preserved while effectively removing noise;The method that the calculating of last median is divided using mean value, this method compare the method that ranking method solves median, reduce the time for calculating median.
Description
Technical field
The present invention relates to a kind of noise filtering method of image, especially a kind of efficient side for eliminating high density salt-pepper noise
Method belongs to the technical field of image preprocessing.
Background technology
Image is often influenced by various noises during acquisition, transmission, reception or processing, such as transmission medium
It is limited with the performance of accepting device, is inevitably present and outwardly and inwardly interferes, i.e., polluted by impulsive noise.And the spiced salt
Noise is exactly a kind of common impulsive noise, it can lead to the degradation of picture quality, and pole is brought to the subsequent processing of image
Big difficulty, such as edge detection.Therefore how to effectively remove the salt-pepper noise of contaminated image becomes one of image procossing
Important technological problems.
It is filtered for the images with salt and pepper noise, many methods for removing salt-pepper noise are suggested, wherein medium filtering
(Median Filter, MF) is widely used due to the simple and certain details hold capacity of its algorithm, but medium filtering
Each pixel in image can be changed, and poor for highdensity salt-pepper noise removal effect.In order to overcome this problem,
People add noise pixel identification link in medium filtering.Wherein noise adaptive fuzzy switch intermediate value (Noise
Adaptive Fuzzy Switching Median, abbreviation NAFSM) filtering give a kind of histogram using noise image
To identify the effective ways of salt-pepper noise.Typically for 8 gray level images, gray value is defined as spiced salt picture for 0 and 255
Element.It is a kind of efficiently fast on the basis of noise adaptive fuzzy switching median filter method for high density the images with salt and pepper noise
The high density salt-pepper noise elimination algorithm of speed is suggested, and the algorithm is in noise reconnaissance phase using in noise adaptive fuzzy switch
Value filtering method is filtering out the salt-pepper noise stage using mean filter, and the spiced salt that this method can efficiently remove high pollution degree is made an uproar
Acoustic image, but the edge details of image cannot be preserved well while filtering out noise.
Invention content
For problems of the prior art and deficiency, the present invention disappears in a kind of efficiently quick high density salt-pepper noise
Except being improved on the basis of algorithm, a kind of efficient method for eliminating high density salt-pepper noise is proposed, to improve elimination high density
The signal-to-noise ratio of salt-pepper noise, and can preferably preserve the edge details of image.
According to technical solution provided by the invention, a kind of efficient method for eliminating high density salt-pepper noise, including following step
Suddenly:
U1,8 gray level image X are converted images into;
U2, to the x in image XijPixel is demarcated, if xijGray value be 0 or 255, then demarcate Fij=1, otherwise
Demarcate Fij=0;
U3, intermediate value filtering, and the method that the calculating of median is divided using mean value are carried out to the salt-pepper noise of calibration;
U4, output filter result image.
In the step U3, to the x of calibrationijSalt-pepper noise carries out intermediate value filtering, includes the following steps:
U3.1, for the spiced salt pixel x in image Xij, define 3 × 3 filter windows centered on the pixel;
U3.2, image Y is defined, is located at the pixel y of (i, j)ij, formula is as follows:
Even FijWhen=1, then yij=mij, and flag Fij=0, if FijWhen=0, then yij=xij.With xijCentered on it is non-
Other eight pixels of center, and the spiced salt pixel that gray value is 0 or 255 is excluded, then calculate the non-spiced salt pixel
Median, the median, that is, mijIf other eight pixels are all spiced salt pixel, median is not calculated.
In the step U3.2, the method that median uses mean value to divide is calculated, is included the following steps:
S1, judge whether the number of element set is 1, if number is 1, which is median, is otherwise walked
Rapid S2;
The mean value M of S2, calculating elements set;
S3, the set H more than or equal to M and the set L less than M are divided by boundary of mean value M;
The number N of S4, set of computations HHWith the number N of set LL;
S5, the number N for judging set LLWhether it is 0, if NLWhen=0, it is the set to take first in set H element
Median, otherwise carry out step S6;
S6, judge NHAnd NLSize, give up the few set of element number, finally to it is obtained set be ranked up method,
The median entirely gathered.
The beneficial effect of the present invention compared with the prior art is:
A kind of efficient method for eliminating high density salt-pepper noise provided by the present invention, first to the salt-pepper noise in image
It is demarcated, intermediate value filtering, and the method that the calculating of median is divided using mean value then is carried out to the salt-pepper noise of calibration.
First, the present invention only filters out the salt-pepper noise of calibration, rather than is filtered to all pixels, therefore the hair
Bright filtering time is solely dependent upon the density of salt-pepper noise;Secondly, intermediate value filtering is carried out to salt-pepper noise, can be gone effectively
The edge details of image are preferably preserved while except noise;Finally, the method that the calculating of median is divided using mean value, the party
Method compares ranking method, reduces the time for calculating median.
Description of the drawings
Fig. 1 is present invention specific implementation flow diagram.
Fig. 2 is the flow diagram for calculating median.
Fig. 3 is the result figure that using the present invention and algorithm tests Lena images before improving.
Fig. 4 is the signal-to-noise ratio comparison result figure that using the present invention and algorithm tests Lena images before improving.
Specific implementation mode
With reference to embodiment and compares attached drawing the present invention is described in further details.
As shown in Figure 1, for present invention specific implementation flow diagram, include the following steps:
U1,8 gray level image X are converted images into;
U2, to the x in image XijPixel is demarcated, if xijGray value be 0 or 255, then demarcate Fij=1, otherwise
Demarcate Fij=0;
U3, intermediate value filtering, and the method that the calculating of median is divided using mean value are carried out to the salt-pepper noise of calibration;
U4, output filter result image.
Wherein in U2 steps, to the x in image XijPixel is demarcated.
For 8 gray level images, the gray value of spiced salt pixel is 0 or 255.Therefore for image X, if some pixel xij
Gray value be 0 or 255, then the pixel may be spiced salt pixel, defined label Fij, formula is as follows:
Work as FijWhen=1, pixel x is indicatedijIt may be spiced salt pixel;Work as FijWhen=0, pixel x is indicatedijFor non-spiced salt picture
Element.
Wherein in U3 steps, intermediate value filtering is carried out to the salt-pepper noise of calibration, is included the following steps:
U3.1, for the spiced salt pixel x in image Xij, define 3 × 3 filter windows centered on the pixel, window
Mouth is as follows:
U3.2, spiced salt pixel currently processed in order to prevent can be determined to other by spiced salt effect pixels to be processed
Adopted image Y is located at the pixel y of (i, j)ij, formula is as follows:
Even FijWhen=1, then yij=mij, spiced salt grey scale pixel value becomes median, and flag Fij=0;If Fij=0
When, then yij=xij, grey scale pixel value is constant.
With xijCentered on non-central location other eight pixels, and exclude gray value be 0 or 255 spiced salt pixel,
Then the median of the non-spiced salt pixel is calculated, the median, that is, mijIf other eight pixels are all spiced salt pixel, do not calculate
Median;
As shown in Fig. 2, to calculate the flow diagram of median, include the following steps:
S1, judge whether the number of element set is 1, if number is 1, which is median, is otherwise walked
Rapid S2;
The mean value M of S2, calculating elements set;
S3, the set H more than or equal to M and the set L less than M are divided by boundary of mean value M;
The number N of S4, set of computations HHWith the number N of set LL;
S5, the number N for judging set LLWhether it is 0, if NLWhen=0, i.e., element is all identical in the set, takes in set H
First element is the median of the set, otherwise carries out step S6;
S6, judge NHAnd NLSize, if NL≥NHWhen, then find from small to large in set L with quick sortA element, which is the median gathered, if NL< NHWhen, then use quick sort in set H from it is small to
It is big to find theA element, the value are the median gathered.
As shown in figure 3, for the result figure tested Lena images using algorithm before the present invention and improvement.In Fig. 3 (a)
For test image Lena;(b) it is plus noise than the images with salt and pepper noise for 60%;(c) it is the result figure improved preceding algorithm and obtained;
(d) it is the obtained result figure of the present invention.As can be seen from the figure algorithm compared with before-improvement of the invention can preferably filter out the spiced salt and make an uproar
Sound, and can preferably preserve the edge details of image.
As shown in figure 4, for the signal-to-noise ratio comparison result tested Lena images using algorithm before the present invention and improvement
Figure.Signal-to-noise ratio PSNR (Peak Signal-to-Noise Ratio) value is used for the quality that objective estimation restores image, PSNR values
It is bigger, indicate that restore image approaches with original image.Salt-pepper noise density changes to 90% from 10%, and one is tested every 10%
It is secondary.It will be seen that when salt-pepper noise density is less than 90% from PSNR test result comparison diagrams, method of the invention is wanted
Significantly better than the method before improvement.
Claims (2)
1. a kind of efficient method for eliminating high density salt-pepper noise, which is characterized in that include the following steps:
U1,8 gray level image X are converted images into;
U2, to the x in image XijPixel is demarcated, if xijGray value be 0 or 255, then demarcate Fij=1, otherwise demarcate
Fij=0;
U3, intermediate value filtering, and the method that the calculating of median is divided using mean value are carried out to the salt-pepper noise of calibration;
U4, output filter result image;
In the step U3, to the x of calibrationijSalt-pepper noise carries out intermediate value filtering, includes the following steps:
U3.1, for the spiced salt pixel x in image Xij, define 3 × 3 filter windows centered on the pixel;
U3.2, image Y is defined, is located at the pixel y of (i, j)ij, formula is as follows:
Even FijWhen=1, then yij=mij, and flag Fij=0, if FijWhen=0, then yij=xij;
With xijCentered on non-central location other eight pixels, and exclude gray value be 0 or 255 spiced salt pixel, then
The median of the non-spiced salt pixel is calculated, the median, that is, mijIf other eight pixels are all spiced salt pixel, centre is not calculated
Value.
2. a kind of efficient method for eliminating high density salt-pepper noise according to claim 1, which is characterized in that the step
U3.2 calculates the method that median uses mean value to divide, and includes the following steps:
S1, judge whether the number of element set is 1, if number is 1, which is median, otherwise carries out step S2;
The mean value M of S2, calculating elements set;
S3, the set H more than or equal to M and the set L less than M are divided by boundary of mean value M;
The number N of S4, set of computations HHWith the number N of set LL;
S5, the number N for judging set LLWhether it is 0, if NLWhen=0, it is in the set to take first in set H element
Between be worth, otherwise carry out step S6;
S6, judge NHAnd NLSize, give up the few set of element number, finally to it is obtained set be ranked up method, to
The median entirely gathered.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710041140.7A CN108320269A (en) | 2017-01-18 | 2017-01-18 | A kind of efficient method for eliminating high density salt-pepper noise |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710041140.7A CN108320269A (en) | 2017-01-18 | 2017-01-18 | A kind of efficient method for eliminating high density salt-pepper noise |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108320269A true CN108320269A (en) | 2018-07-24 |
Family
ID=62891705
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710041140.7A Pending CN108320269A (en) | 2017-01-18 | 2017-01-18 | A kind of efficient method for eliminating high density salt-pepper noise |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108320269A (en) |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7110612B1 (en) * | 2001-10-11 | 2006-09-19 | Pixelworks, Inc. | Weighted absolute difference based noise reduction method and apparatus |
US7203339B1 (en) * | 2003-06-26 | 2007-04-10 | The United States Of America As Represented By The Secretary Of The Navy | Enhancing two-dimensional contrast and range images rendered from three-dimensional streak tube imaging lidar (STIL) data |
CN103400357A (en) * | 2013-08-23 | 2013-11-20 | 闽江学院 | Method for removing salt-pepper noises in images |
CN103679732A (en) * | 2013-12-12 | 2014-03-26 | 西安建筑科技大学 | Noise-detection-based high density impulse noise self-adaptive filtering algorithm |
CN103761707A (en) * | 2013-12-20 | 2014-04-30 | 浙江大学 | Average filtering method eliminating image impulse noise fast and efficiently |
US20140133774A1 (en) * | 2012-11-09 | 2014-05-15 | Industrial Technology Research Institute | Image processor and image dead pixel detection method thereof |
CN103871034A (en) * | 2014-03-22 | 2014-06-18 | 四川大学 | Self-adapting filtering method for salt and pepper noise of image |
CN104167005A (en) * | 2014-07-07 | 2014-11-26 | 浙江大学 | Salt and pepper noise filtering method based on similar function with better self-adaptation, denoising and detail protection capabilities |
CN104915938A (en) * | 2015-07-02 | 2015-09-16 | 中国人民解放军国防科学技术大学 | Restoration method for image with high density impulse noise pollution |
CN105654442A (en) * | 2015-12-31 | 2016-06-08 | 大连理工大学 | Noise elimination method for impact noise image |
CN105719257A (en) * | 2016-01-28 | 2016-06-29 | 河南师范大学 | Method for removing super-high-density salt-and-pepper noises of image |
-
2017
- 2017-01-18 CN CN201710041140.7A patent/CN108320269A/en active Pending
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7110612B1 (en) * | 2001-10-11 | 2006-09-19 | Pixelworks, Inc. | Weighted absolute difference based noise reduction method and apparatus |
US7203339B1 (en) * | 2003-06-26 | 2007-04-10 | The United States Of America As Represented By The Secretary Of The Navy | Enhancing two-dimensional contrast and range images rendered from three-dimensional streak tube imaging lidar (STIL) data |
US20140133774A1 (en) * | 2012-11-09 | 2014-05-15 | Industrial Technology Research Institute | Image processor and image dead pixel detection method thereof |
CN103400357A (en) * | 2013-08-23 | 2013-11-20 | 闽江学院 | Method for removing salt-pepper noises in images |
CN103679732A (en) * | 2013-12-12 | 2014-03-26 | 西安建筑科技大学 | Noise-detection-based high density impulse noise self-adaptive filtering algorithm |
CN103761707A (en) * | 2013-12-20 | 2014-04-30 | 浙江大学 | Average filtering method eliminating image impulse noise fast and efficiently |
CN103871034A (en) * | 2014-03-22 | 2014-06-18 | 四川大学 | Self-adapting filtering method for salt and pepper noise of image |
CN104167005A (en) * | 2014-07-07 | 2014-11-26 | 浙江大学 | Salt and pepper noise filtering method based on similar function with better self-adaptation, denoising and detail protection capabilities |
CN104915938A (en) * | 2015-07-02 | 2015-09-16 | 中国人民解放军国防科学技术大学 | Restoration method for image with high density impulse noise pollution |
CN105654442A (en) * | 2015-12-31 | 2016-06-08 | 大连理工大学 | Noise elimination method for impact noise image |
CN105719257A (en) * | 2016-01-28 | 2016-06-29 | 河南师范大学 | Method for removing super-high-density salt-and-pepper noises of image |
Non-Patent Citations (3)
Title |
---|
K S SRINIVASAN ET AL: ""A new fast and efficient decision based algorithm for removal of high-density impulse noise", 《IEEE SIGNAL PROCESSING LETTERS》 * |
吕宗伟等: "一种高效快速的高密度椒盐噪声消除算法", 《电子学报》 * |
张丽等: "均值加速的快速中值滤波算法", 《清华大学学报(自然科学版)》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Raza et al. | High density salt and pepper noise removal through decision based partial trimmed global mean filter | |
Gao | An adaptive median filtering of salt and pepper noise based on local pixel distribution | |
CN101655977B (en) | Method for eliminating image impulse noise based on differential image detection and filtration by multiple windows | |
Agrawal et al. | A survey of linear and non-linear filters for noise reduction | |
Zhang et al. | Modified adaptive median filtering | |
Mohapatra et al. | Histogram equalization and noise removal process for enhancement of image | |
Tang et al. | Improved adaptive median filtering for structured light image denoising | |
Yinyu et al. | A Study on Image Restoration Algorithm in Random-valued Impulse Noise Environment | |
Mehta et al. | Comparative analysis of median filter and adaptive filter for impulse noise–a review | |
Mohammed | An improved median filter based on efficient noise detection for high quality image restoration | |
Jayasree et al. | An efficient mixed noise removal technique from gray scale images using noisy pixel modification technique | |
CN108320269A (en) | A kind of efficient method for eliminating high density salt-pepper noise | |
Krishna et al. | Removal of high density salt and pepper noise through modified decision based unsymmetric trimmed median filter | |
Smolka | Soft switching technique for impulsive noise removal in color images | |
CN101197934B (en) | Method and device for reducing noise between frames | |
Coumar et al. | Image restoration using filters and image quality assessment using reduced reference metrics | |
Bansal et al. | New methodology for SP noise removal in digital image processing | |
Lakshmi et al. | Impulse Noise Removal Inimages Using Modified Trimmed Median Filter: Matlab Implementation And Comparitive Study | |
Langampol et al. | Applied Switching Bilateral Filter for Color Images under Mixed Noise | |
Palabaş et al. | Adaptive fuzzy filter combined with median filter for reducing intensive salt and pepper noise in gray level images | |
Prasad | Color and gray scale image denoising using modified Decision Based unsymmetric Trimmed Median Filter | |
Chen et al. | An impulse noise reduction method by adaptive pixel-correlation | |
Bhatia et al. | High density salt and pepper noise removal through improved adaptive median filter | |
Khan et al. | A novel algorithm for removal of noise from X-Ray images | |
Agrawal et al. | A Novel Weighted Median Switching Filter for Denoising Corrupted Images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180724 |
|
WD01 | Invention patent application deemed withdrawn after publication |