CN102622597B - Adaptive quadrature intermediate value mixed filtering method - Google Patents
Adaptive quadrature intermediate value mixed filtering method Download PDFInfo
- Publication number
- CN102622597B CN102622597B CN201110031477.2A CN201110031477A CN102622597B CN 102622597 B CN102622597 B CN 102622597B CN 201110031477 A CN201110031477 A CN 201110031477A CN 102622597 B CN102622597 B CN 102622597B
- Authority
- CN
- China
- Prior art keywords
- value
- condition
- pixel
- image
- pixels
- 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.)
- Expired - Fee Related
Links
Landscapes
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
A kind of adaptive quadrature intermediate value mixed filtering method, belongs to image recognition technology field. Object of the present invention, in retaining original image more details, has been removed again the adaptive quadrature intermediate value mixed filtering method of picture noise well. Step of the present invention is: take out all the other four pixels at two orthogonal directions taking center pixel respectively as core, be divided into into two groups: I group pixel groups and II group pixel groups; After each pixel of picture is processed equally, will obtain an image after smothing filtering. The present invention is mainly used in the smoothing processing of image in image recognition processing process. In image more details can be retained in image filtering process, remove better picture noise.
Description
Technical field
The invention belongs to image recognition technology field.
Background technology
Existing filtering algorithm has also retained picture noise in retaining image detail, is unfavorable for the identification of image detailWith judgement, and existing filtering algorithm is considered image detail except the noise time and is also removed or obfuscation.
Summary of the invention
Object of the present invention, in retaining original image more details, has been removed again the adaptive of picture noise wellShould orthogonal intermediate value mixed filtering method.
Step of the present invention is:
A, take out all the other four pixels at two orthogonal directions taking center pixel respectively as core, be divided into into two groups: I groupPixel groups and II group pixel groups;
B, condition one: I class mean is less than the maximum of five pixel gray values in I group, and be greater than five pixels of I groupThe minimum of a value of some gray value;
C, condition two: II class mean is less than the maximum of five pixel gray values in II group, and be greater than five of II groupsThe minimum of a value of pixel gray value;
If d condition one meets, condition two does not meet, and the gray value of the center pixel of nine pixels of getting is I with regard to valueThe intermediate value of group;
If e condition two meets, condition one does not meet, and the gray value of the center pixel of nine pixels of getting with regard to value isThe intermediate value of II group;
If f condition one does not meet with condition two or condition one all meets with condition two, the center of nine pixels of gettingThe gray value of pixel is the mean value of I group and II class mean with regard to value;
G, each pixel of picture is processed equally after, will obtain an image after smothing filtering.
The present invention is mainly used in the smoothing processing of image in image recognition processing process. Can in image filtering process, protectWhen staying image more details, remove better picture noise.
Brief description of the drawings
Fig. 1 is original image;
Fig. 2 adds the images with salt and pepper noise;
Fig. 3 is mean filter image;
Fig. 4 is adaptive median filter image;
Fig. 5 is medium filtering image;
Fig. 6 is adaptive quadrature intermediate value mixed filtering image of the present invention;
Fig. 7 is that the present invention gets orthogonal neighbor pixel schematic diagram anyhow with central pixel point;
Fig. 8 is that the present invention gets orthogonal neighbor pixel schematic diagram with central pixel point skewed crossing.
Detailed description of the invention
Step of the present invention is:
A, take out all the other four pixels at two orthogonal directions taking center pixel respectively as core, be divided into into two groups: I groupPixel groups and II group pixel groups;
B, condition one: I class mean is less than the maximum of five pixel gray values in I group, and be greater than five pixels of I groupThe minimum of a value of some gray value;
C, condition two: II class mean is less than the maximum of five pixel gray values in II group, and be greater than five of II groupsThe minimum of a value of pixel gray value;
If d condition one meets, condition two does not meet, and the gray value of the center pixel of nine pixels of getting is I with regard to valueThe intermediate value of group;
If e condition two meets, condition one does not meet, and the gray value of the center pixel of nine pixels of getting with regard to value isThe intermediate value of II group;
If f condition one does not meet with condition two or condition one all meets with condition two, the center of nine pixels of gettingThe gray value of pixel is the mean value of I group and II class mean with regard to value;
G, each pixel of picture is processed equally after, will obtain an image after smothing filtering.
Below invention is done to concrete description:
Shown in Fig. 7 and Fig. 8, nine grids represent nine adjacent pixels of left and right in a certain picture, taking center pixel as coreThe heart, by taking out all the other four pixels at two orthogonal directions respectively shown in figure, is divided into into two groups shown in Fig. 7 and Fig. 8.
In I group pixel groups, will after the gray value summation of five pixels, average and be defined as the intermediate value of I group pixel.
Condition one: I class mean is less than the maximum of five pixel gray values in I group, and be greater than five pixels of I groupThe minimum of a value of gray value.
In like manner in II group pixel groups, be defined as averaging after the gray value summation of five pixels II group pixel inValue.
Condition two: II class mean is less than the maximum of five pixel gray values in II group, and be greater than five pictures of II groupThe minimum of a value of vegetarian refreshments gray value.
If condition one meets, condition two does not meet, and the gray value of the center pixel of nine pixels of getting is I group with regard to valueIntermediate value.
If condition two meets, condition one does not meet, and the gray value of the center pixel of nine pixels of getting is II with regard to valueThe intermediate value of group.
If condition one does not meet with condition two or condition one all meets with condition two, the middle imago of nine pixels of gettingThe gray value of element is the mean value of I group and II class mean with regard to value.
According to above-mentioned processing mode, after each pixel of picture is processed equally, will obtain a level and smooth filterImage after ripple, this method is defined as adaptive quadrature intermediate value mixed filtering method.
Represent that one exists processed, centerThe subimage at place.
With SxyFor example illustrates the implementation procedure of this algorithm.
Extract two groups of pixels by mode shown in Fig. 7 and Fig. 8、, and do to give a definition:
RepresentMiddle minimum luminance value
RepresentMiddle maximum brightness value
RepresentIn brightness intermediate value
RepresentMiddle minimum luminance value
RepresentMiddle maximum brightness value
RepresentIn brightness intermediate value
RepresentLocate final output valve
If Juz1 is true, Juz2 vacation
=;
If Juz1 vacation, Juz2 is true
=; Otherwise=1/2(+)
Claims (1)
1. an adaptive quadrature intermediate value mixed filtering method, is characterized in that:
A, take out all the other four pixels at two orthogonal directions taking center pixel respectively as core, be divided into into two groups: I organizes pixelGroup and II group pixel groups;
B, condition one: I class mean is less than the maximum of five pixel gray values in I group, and be greater than five pixel ashes of I groupThe minimum of a value of degree value;
C, condition two: II class mean is less than the maximum of five pixel gray values in II group, and be greater than five pixels of II groupThe minimum of a value of some gray value;
If d condition one meets, condition two does not meet, and the gray value of the center pixel of nine pixels of getting is that I organizes with regard to valueIntermediate value;
If e condition two meets, condition one does not meet, and the gray value of the center pixel of nine pixels of getting is II group with regard to valueIntermediate value;
If f condition one does not meet with condition two or condition one all meets with condition two, the center pixel of nine pixels of gettingGray value be I group and the mean value of II class mean with regard to value;
G, each pixel of picture is processed equally after, will obtain an image after smothing filtering.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110031477.2A CN102622597B (en) | 2011-01-29 | 2011-01-29 | Adaptive quadrature intermediate value mixed filtering method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110031477.2A CN102622597B (en) | 2011-01-29 | 2011-01-29 | Adaptive quadrature intermediate value mixed filtering method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102622597A CN102622597A (en) | 2012-08-01 |
CN102622597B true CN102622597B (en) | 2016-05-04 |
Family
ID=46562507
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201110031477.2A Expired - Fee Related CN102622597B (en) | 2011-01-29 | 2011-01-29 | Adaptive quadrature intermediate value mixed filtering method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102622597B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101094312A (en) * | 2006-06-20 | 2007-12-26 | 西北工业大学 | Self-adapting method for filtering image with edge being retained |
CN101388113A (en) * | 2008-10-24 | 2009-03-18 | 北京航空航天大学 | Star map image rapid denoising method |
CN101425176A (en) * | 2008-12-09 | 2009-05-06 | 中国科学院长春光学精密机械与物理研究所 | Image wavelet de-noising method based on median filter |
CN101944230A (en) * | 2010-08-31 | 2011-01-12 | 西安电子科技大学 | Multi-scale-based natural image non-local mean noise reduction method |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7317842B2 (en) * | 2003-10-30 | 2008-01-08 | Samsung Electronics Co., Ltd. | Global and local statistics controlled noise reduction system |
US7899248B2 (en) * | 2007-08-30 | 2011-03-01 | Seiko Epson Corporation | Fast segmentation of images |
-
2011
- 2011-01-29 CN CN201110031477.2A patent/CN102622597B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101094312A (en) * | 2006-06-20 | 2007-12-26 | 西北工业大学 | Self-adapting method for filtering image with edge being retained |
CN101388113A (en) * | 2008-10-24 | 2009-03-18 | 北京航空航天大学 | Star map image rapid denoising method |
CN101425176A (en) * | 2008-12-09 | 2009-05-06 | 中国科学院长春光学精密机械与物理研究所 | Image wavelet de-noising method based on median filter |
CN101944230A (en) * | 2010-08-31 | 2011-01-12 | 西安电子科技大学 | Multi-scale-based natural image non-local mean noise reduction method |
Non-Patent Citations (2)
Title |
---|
改进自适应中值滤波的图像去噪;肖蕾等;《激光杂志》;20090415;第30卷(第2期);第44-46页 * |
用于图像处理的自适应中值滤波;张旭明等;《计算机辅助设计与图形学学报》;20050220;第17卷(第2期);第295-299页 * |
Also Published As
Publication number | Publication date |
---|---|
CN102622597A (en) | 2012-08-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liu et al. | Joint demosaicing and denoising with self guidance | |
CN106096668B (en) | The recognition methods and identifying system of watermarked image | |
Fan et al. | Homomorphic filtering based illumination normalization method for face recognition | |
CN102222328B (en) | Edge-preserving self-adaptive weighted filtering method for natural scene images | |
JP5767064B2 (en) | Image edge enhancement method | |
CN111179189B (en) | Image processing method and device based on generation of countermeasure network GAN, electronic equipment and storage medium | |
CN106815821A (en) | The denoising method and device of near-infrared image | |
Song et al. | Decolorization: Is rgb2gray () out? | |
CN102509294B (en) | Single-image-based global depth estimation method | |
TW200917853A (en) | Noise reduced color image using panchromatic image | |
CN105469353B (en) | The embedding grammar and device and extracting method and device of watermarking images | |
CN107767356B (en) | Image processing method and device | |
CN104657941B (en) | A kind of image border self-adapting enhancement method and device | |
JPWO2002067589A1 (en) | Image processing system, image processing method, and image processing program | |
CN105389778B (en) | A kind of image super-resolution rebuilding method and device based on dictionary pattern matching | |
CN109003247B (en) | Method for removing color image mixed noise | |
CN102622597B (en) | Adaptive quadrature intermediate value mixed filtering method | |
CN104143203A (en) | Image editing and communication method | |
Zhao et al. | Single depth image super-resolution with multiple residual dictionary learning and refinement | |
CN105872317B (en) | A kind of video image filtering method and device based on time domain | |
CN108550119B (en) | Image denoising method combined with edge information | |
CN111652809A (en) | Infrared image noise suppression method for enhancing details | |
CN106897975B (en) | Image denoising method for hypercube particle calculation | |
CN106887024B (en) | The processing method and processing system of photo | |
CN103077396A (en) | Method and device for extracting vector space feature points of color image |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160504 Termination date: 20200129 |
|
CF01 | Termination of patent right due to non-payment of annual fee |