CN109448014A - A kind of image information thinning method based on subgraph - Google Patents
A kind of image information thinning method based on subgraph Download PDFInfo
- Publication number
- CN109448014A CN109448014A CN201811223265.2A CN201811223265A CN109448014A CN 109448014 A CN109448014 A CN 109448014A CN 201811223265 A CN201811223265 A CN 201811223265A CN 109448014 A CN109448014 A CN 109448014A
- Authority
- CN
- China
- Prior art keywords
- subgraph
- image
- tseg
- gray level
- histogram
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
Abstract
The present invention relates to a kind of image information thinning method based on subgraph.Include the following steps: step 1, input gray level image；Step 2, grey level histogram is calculated；Step 3, global gray level ratio is calculated；Step 4, dividing sub-picture threshold value is calculated；Step 5, grey level histogram is trimmed；Step 6, dividing sub-picture；Step 7, density function is calculated；Step 8, distribution function is calculated；Step 9, histogram transfer function is constructed；Step 10, equalization processing；Step 11, it merges subgraph and exports.The local message of image can be refined, can be applied to the fields such as the enhancing processing of fuzzy digit image.
Description
Technical field
The present invention relates to a kind of digital image processing field, specifically a kind of image information refinement side based on subgraph
Method.
Background technique
Much since the visual effect of the influence image taking of scene condition is bad, this just needs image thinning technology to change
Certain features of target object in the visual effect of philanthropist, such as prominent image, the feature that object is extracted from digital picture
Parameter etc., these are all conducive to the identification, tracking and understanding to target in image.Image thinning processing to the effect that protrudes
Unwanted information is weakened or removed in interested part in image.Useful information is set to be strengthened in this way, to obtain one kind
More practical image is converted into a kind of image for being more suitable for people or machine is analyzed and processed.Currently used related side
Method includes gray level correction, greyscale transformation, histogram modification, image smoothing and sharpening etc., however, in algorithm complexity and refinement
Effect etc. is still to be improved.
Summary of the invention
The present invention provides a kind of, and the image information thinning method based on subgraph is distinguished using by the principle of dividing sub-picture
Using the methods of histogram, density function and distribution function, calculation amount is small, and image thinning effect is good.
Technical solution used by target to realize the present invention is: method the following steps are included:
Step 1: input gray level image I；
Step 2: calculating the grey level histogram H of gray level image I；
Step 3: calculating the global gray level ratio Lg of gray level image；
Step 4: calculating dividing sub-picture threshold value Tseg, Tseg=G (1-Lg)；
Step 5: information refinement excessively causes the information of gray level image I to be lost in order to prevent, needs to grey level histogram H
It is trimmed, when H (i) is greater than trimming threshold value Tc, H (i)=Tc, the grey level histogram after trimming is denoted as Hc；
Step 6: gray level image I being divided into I with dividing sub-picture threshold value TsegLAnd IHTwo subgraphs, wherein subgraph IL's
Gray scale value range is [0, Tseg], subgraph IHGray scale value range be [Tseg+1, G-1]；
Step 7: calculating separately subgraph ILWith subgraph IHDensity function PLAnd PH, wherein PL=Hc/ML, PH=Hc/MH, ML
And MHRespectively subgraph ILWith subgraph IHPixel quantity；
Step 8: calculating separately subgraph ILWith subgraph IHCumulative Distribution Function CLAnd CH；
Step 9: the Cumulative Distribution Function C being utilized respectively in step 8LAnd CHConstruct subgraph ILWith subgraph IHHistogram pass
Defeated function TFLAnd TFH；
Step 10: respectively to the histogram transfer function TF in step 9LAnd TFHEqualization processing is carried out, is respectively obtained
The beggar that weighs schemes ILQAnd IHQ；
Step 11: the equalization subgraph I in fusion steps 10LQAnd IHQ, obtain information refinement image IQAnd it exports.
Global gray level ratio Lg described in step 3 is realized by formula (1):
In formula (1), H (i) is the quantity for the pixel that gray value is i in gray level image I, and G is the gray scale etc. of gray level image I
Grade.
Trimming threshold value Tc described in step 5 is calculated by formula (2)；
Subgraph I is calculated separately described in step 8LWith subgraph IHCumulative Distribution Function CLAnd CHIt is calculated by formula (3)；
Cumulative Distribution Function C described in step 9LAnd CHCumulative Distribution Function CLAnd CH, it is calculated respectively by formula (4),
The beneficial effects of the present invention are: can refine to the local message of image, fuzzy digit figure can be applied to
The fields such as the enhancing processing of picture.
Detailed description of the invention
Fig. 1 is overall process flow figure of the invention.
Specific embodiment
It describes the specific embodiments of the present invention in detail with reference to the accompanying drawing.
In Fig. 1,101 be input gray level image step, and 102 be to calculate grey level histogram step, and 103 be to calculate global gray scale
Ratio step, 104 calculate dividing sub-picture threshold steps, and 105 trimming ash top histogram steps, 106 be dividing sub-picture step, and 107
It is to calculate density function step, 108 be to calculate distribution function step, and 109 be construction histogram transfer function step, and 110 is balanced
Change processing step, 111 are fusion subgraphs and export step.
Specific embodiment process is as follows:
Step 101: input gray level image I；
Step 102: calculating the grey level histogram H of gray level image I；
Step 103: calculate the global gray level ratio Lg of image:
In formula, H (i) is the quantity for the pixel that gray value is i in gray level image I, and G is the tonal gradation of gray level image I；
Step 104: calculating dividing sub-picture threshold value Tseg, Tseg=G (1-Lg)；
Step 105: information refinement excessively causes the information of gray level image I to be lost in order to prevent, needs to intensity histogram
Figure H is trimmed, and when H (i) is greater than trimming threshold value Tc, H (i)=Tc, the grey level histogram after trimming is denoted as Hc, wherein it repairs
Threshold value Tc is cut to be calculate by the following formula；
Step 106: gray level image I being divided into I with dividing sub-picture threshold value TsegLAnd IHTwo subgraphs, wherein subgraph IL
Gray scale value range be [0, Tseg], subgraph IHGray scale value range be [Tseg+1, G-1]；
Step 107: calculating separately subgraph ILWith subgraph IHDensity function PLAnd PH, wherein PL=Hc/ML, PH=Hc/MH,
MLAnd MHRespectively subgraph ILWith subgraph IHPixel quantity；
Step 108: calculating separately subgraph ILWith subgraph IHDistribution function CLAnd CH；
Step 109: the distribution function C being utilized respectively in step 108LAnd CHConstruct subgraph ILWith subgraph IHHistogram pass
Defeated function TFLAnd TFH；
TFL=Tseg × CL
TFH=(Tseg+1)+(G-Tseg+1) × CH
Step 110: respectively to the histogram transfer function TF in step 109LAnd TFHEqualization processing is carried out, is respectively obtained
Equalize subgraph ILQAnd IHQ；
Step 111: the equalization subgraph I in fusion steps 110LQAnd IHQ, obtain information refinement image IQAnd it exports.
Claims (6)
1. a kind of image information thinning method based on subgraph, it is characterised in that the following steps are included:
Step 1: input gray level image I；
Step 2: calculating the grey level histogram H of gray level image I；
Step 3: calculating the global gray level ratio Lg of gray level image；
Step 4: calculating dividing sub-picture threshold value Tseg, Tseg=G (1-Lg)；
Step 5: grey level histogram H is trimmed, when H (i) is greater than trimming threshold value Tc, H (i)=Tc, the gray scale after trimming
Histogram is denoted as Hc；
Step 6: gray level image I being divided into I with dividing sub-picture threshold value TsegLAnd IHTwo subgraphs, wherein subgraph ILGray scale
Value range is [0, Tseg], subgraph IHGray scale value range be [Tseg+1, G-1]；
Step 7: calculating separately subgraph ILWith subgraph IHDensity function PLAnd PH；
Step 8: calculating separately subgraph ILWith subgraph IHDistribution function CLAnd CH；
Step 9: the distribution function C being utilized respectively in step 8LAnd CHConstruct subgraph ILWith subgraph IHHistogram transfer function TFL
And TFH；
Step 10: respectively to the histogram transfer function TF in step 9LAnd TFHEqualization processing is carried out, equalization is respectively obtained
Subgraph ILQAnd IHQ；
Step 11: the equalization subgraph I in fusion steps 10LQAnd IHQ, obtain information refinement image IQAnd it exports.
2. a kind of image information thinning method based on subgraph according to claim 1, it is characterised in that described in step 3
Global gray level ratio Lg is calculated by formula (1):
In formula (1), H (i) is the quantity for the pixel that gray value is i in gray level image I, and G is the tonal gradation of gray level image I.
3. a kind of image information thinning method based on subgraph according to claim 1, it is characterised in that described in step 5
Threshold value Tc is trimmed, is calculated by formula (2):
4. a kind of image information thinning method based on subgraph according to claim 1, it is characterised in that described in step 7
Density function PLAnd PH, PL=Hc/ML, PH=Hc/MH, MLAnd MHRespectively subgraph ILWith subgraph IHPixel quantity.
5. a kind of image information thinning method based on subgraph according to claim 1, it is characterised in that described in step 8
Cumulative Distribution Function CLAnd CH, it is calculated by formula (3):
6. a kind of image information thinning method based on subgraph according to claim 1, it is characterised in that described in step 9
Histogram transfer function TFLAnd TFH, it is calculated by formula (4):
TFL=Tseg × CL
TFH=(Tseg+1)+(G-Tseg+1) × CH。 (4)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811223265.2A CN109448014B (en) | 2018-10-19 | 2018-10-19 | Image information thinning method based on subgraph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811223265.2A CN109448014B (en) | 2018-10-19 | 2018-10-19 | Image information thinning method based on subgraph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109448014A true CN109448014A (en) | 2019-03-08 |
| CN109448014B CN109448014B (en) | 2021-04-30 |
Family
ID=65547548
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811223265.2A Active CN109448014B (en) | 2018-10-19 | 2018-10-19 | Image information thinning method based on subgraph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109448014B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101916371A (en) * | 2010-09-01 | 2010-12-15 | 北京工业大学 | Method for illuminating/normalizing image and method for identifying image by using same |
| CN103440635A (en) * | 2013-09-17 | 2013-12-11 | 厦门美图网科技有限公司 | Learning-based contrast limited adaptive histogram equalization method |
| CN103606137A (en) * | 2013-11-13 | 2014-02-26 | 天津大学 | Histogram equalization method for maintaining background and detail information |
| CN105320946A (en) * | 2015-11-03 | 2016-02-10 | 盐城工学院 | MATLAB based fingerprint identification method |
| CN106127791A (en) * | 2016-07-06 | 2016-11-16 | 武汉大学 | A kind of contour of building line drawing method of aviation remote sensing image |
-
2018
- 2018-10-19 CN CN201811223265.2A patent/CN109448014B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101916371A (en) * | 2010-09-01 | 2010-12-15 | 北京工业大学 | Method for illuminating/normalizing image and method for identifying image by using same |
| CN103440635A (en) * | 2013-09-17 | 2013-12-11 | 厦门美图网科技有限公司 | Learning-based contrast limited adaptive histogram equalization method |
| CN103606137A (en) * | 2013-11-13 | 2014-02-26 | 天津大学 | Histogram equalization method for maintaining background and detail information |
| CN105320946A (en) * | 2015-11-03 | 2016-02-10 | 盐城工学院 | MATLAB based fingerprint identification method |
| CN106127791A (en) * | 2016-07-06 | 2016-11-16 | 武汉大学 | A kind of contour of building line drawing method of aviation remote sensing image |
Non-Patent Citations (1)
| Title |
|---|
| 胡琼: ""基于直方图分割的彩色图像增强算法"", 《中国图象图形学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109448014B (en) | 2021-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101729911B (en) | Multi-view image color correction method based on visual perception | |
| Dai et al. | Single underwater image restoration by decomposing curves of attenuating color | |
| CN107705254B (en) | City environment assessment method based on street view | |
| CN102750674A (en) | Video image defogging method based on self-adapting allowance | |
| Zhao et al. | Single image fog removal based on local extrema | |
| CN109871845B (en) | Certificate image extraction method and terminal equipment | |
| CN105976337B (en) | A kind of image defogging method based on intermediate value guiding filtering | |
| CN105046677A (en) | Enhancement processing method and apparatus for traffic video image | |
| CN103226824B (en) | Maintain the video Redirectional system of vision significance | |
| CN107895379A (en) | The innovatory algorithm of foreground extraction in a kind of video monitoring | |
| CN102208101A (en) | Self-adaptive linearity transformation enhancing method of infrared image | |
| CN100571404C (en) | Skin color signal correcting method | |
| CN103106796A (en) | Vehicle detection method and device of intelligent traffic surveillance and control system | |
| CN109377464A (en) | A kind of Double plateaus histogram equalization method and its application system of infrared image | |
| Li et al. | Underwater image enhancement based on dehazing and color correction | |
| Wang et al. | An efficient method for image dehazing | |
| CN110298796A (en) | Based on the enhancement method of low-illumination image for improving Retinex and Logarithmic image processing | |
| CN106778777A (en) | A kind of vehicle match method and system | |
| CN103136530A (en) | Method for automatically recognizing target images in video images under complex industrial environment | |
| CN106296626B (en) | A kind of night video enhancement method based on gradient fusion | |
| CN112750128B (en) | Image semantic segmentation method, device, terminal and readable storage medium | |
| CN109448014A (en) | A kind of image information thinning method based on subgraph | |
| CN110399886B (en) | Screen image JND model construction method | |
| CN108537823A (en) | Moving target detecting method based on mixed Gauss model | |
| CN111369477A (en) | Method for pre-analysis and tool self-adaptation of video recovery task |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |