CN103763460B - Image sharpening method - Google Patents
Image sharpening method Download PDFInfo
- Publication number
- CN103763460B CN103763460B CN201410025563.6A CN201410025563A CN103763460B CN 103763460 B CN103763460 B CN 103763460B CN 201410025563 A CN201410025563 A CN 201410025563A CN 103763460 B CN103763460 B CN 103763460B
- Authority
- CN
- China
- Prior art keywords
- decision
- point
- threshold
- sharpening
- gtg difference
- 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.)
- Active
Links
Landscapes
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
The invention relates to the field of image processing, in particular to an image sharpening method. The method comprises the steps that eight adjacent points around a judging point are selected, wherein the judging point and the adjacent points are all pixel points in an image; gray scale difference values between the adjacent points and the judging point are calculated respectively; comparison is carried out between the gray scale difference values and a preset sharpening judging threshold value, and the number of the corresponding adjacent points with the gray scale difference values larger than the preset sharpening judging threshold value is accumulated as a judging value; whether the judging point is an edge point or not is judged according to the judging value, and sharpening processing is carried out on the judging point if the judging point is the edge point. According to the image sharpening method, when image sharpening is carried out, only the judging point and the eight adjacent pixel points are selected for carrying out plus, minus and shift operation on gray scales, occupied logic resources are small, and power consumption is reduced.
Description
Technical field
The present invention relates to image processing field, more particularly to a kind of image sharpening method.
Background technology
Image sharpening is to strengthen image border, allows a kind of method that fuzzy image becomes apparent from, the realization of image sharpening
Method has a lot, such as the zero crossings algorithm of the gradient algorithm of first derivative, second dervative, for what is projected in piece image
Marginal zone, its Grad is larger；Little in smooth region Grad, for gray level is the region of constant, Grad is zero, and it two
There is zero crossing at marginal point in order derivative, i.e. the second dervative of marginal point both sides takes contrary sign, and gradient algorithm is exactly to utilize one
The method of order derivative detected edge points, zero crossings algorithm is exactly come detected edge points by second dervative.
If the colored image sharpening method for showing of patent TFT-LCD is using five points with a line, on the basis of central point,
Calculate the gray scale difference value of and each color component of this point at remaining 4 points, then calculate sharpening using sharpening factor computing formula
The factor, then calculates the sharpening enhancement factor and zoom factor, three's value is multiplied with the gray scale difference value of color component and obtains ash
Degree incrementss, this gray scale increments are multiplied with energy dispersion electric-wave filter matrix obtain last color component gray scale increase again
Amount, the gray scale incrementss of each color component are added on the corresponding color component of each pixel.
However, in existing conventional algorithm, it be all to carry out convolution algorithm using operator template to calculate figure that it is implemented
As marginal point, follow-up Edge contrast is carried out, this needs substantial amounts of logical resource when realizing in driver IC.
The content of the invention
It is an object of the invention to propose a kind of image sharpening method, resources occupation rate can be reduced, save power consumption.
It is that, up to this purpose, the present invention is employed the following technical solutions：
A kind of image sharpening method, including：
Eight consecutive points around step a, selection decision-point, the decision-point and consecutive points are all the pixels in image
Point；
Step b, the consecutive points are calculated respectively with the GTG difference of the decision-point；
Step c, by the GTG difference with it is default sharpening decision threshold make comparisons, will be greater than the sharpening decision threshold
The number of the corresponding consecutive points of GTG difference add up as decision content；
Step d, judge whether the decision-point is marginal point according to decision content, if so, then the decision-point carried out sharp
Change is processed.
Wherein, the decision-point is sharpened specially：By the color component of the decision-point plus or minus described
Sharpen decision threshold.
Wherein, it is described to judge whether the decision-point is that marginal point is specially according to decision content：Judge that the decision content is
It is no within default value range, if so, then the decision-point be marginal point.
Wherein, the GTG difference includes single color GTG difference and trichroism GTG difference, the sharpening decision threshold bag
First threshold and Second Threshold are included, the decision content includes the first decision content and the second decision content.
Wherein, when the GTG difference is single color GTG difference, step c, step d are specially：
Step c1, the single color GTG difference is made comparisons with default first threshold, will be greater than the first threshold
The number of the corresponding consecutive points of single color GTG difference add up as the first decision content；
Step d1, judge first decision content whether within default value range, if so, then the decision-point is side
Edge point.
Wherein, when the GTG difference is trichroism GTG difference, step c, step d are specially：
Step c2, the trichroism GTG difference is made comparisons with default Second Threshold, by trichroism GTG difference at least
Dichromatism GTG difference adds up to sentence for second more than the number of the corresponding consecutive points of trichroism GTG difference of the Second Threshold
Definite value；
Step d2, judge second decision content whether within default value range, if so, then the decision-point is side
Edge point.
Wherein, in step a, the number of consecutive points is eight, eight consecutive points respectively positioned at the decision-point it is upper,
Under, left and right, upper left, upper right, lower-left and the direction of bottom right eight.
Wherein, the default value range is that eight are less than more than or equal to two, when the decision content of the decision-point is less than two,
The decision-point is non-edge point；When the decision content of the decision-point is equal to eight, the decision-point is isolated noise spot.
Wherein, also include after step d：
If step e, the decision-point are not marginal points, whether it is judged as isolated noise spot, if so, then to noise spot
Do smoothing processing.
Beneficial effects of the present invention are：A kind of image sharpening method, including：Eight consecutive points around decision-point are chosen,
The decision-point and consecutive points are all the pixels in image；The grey jump of the consecutive points and the decision-point is calculated respectively
Value；The GTG difference is made comparisons with default sharpening decision threshold, the GTG difference of the sharpening decision threshold is will be greater than
The number of the corresponding consecutive points adds up as decision content；Judge whether the decision-point is marginal point according to decision content, if so,
Then it is sharpened process to the decision-point, the present invention only needs to choose decision-point and adjacent with decision-point in image sharpening
Eight pixels do plus-minus and the shift operation of GTG, and the logical resource of occupancy is less, saves power consumption.
Description of the drawings
Fig. 1 is the image sharpening method flow chart that the specific embodiment of the invention is provided.
Fig. 2 is that the consecutive points of the decision-point that the specific embodiment of the invention is provided choose schematic diagram.
Fig. 3 is the pending original image that the specific embodiment of the invention is provided.
Fig. 4 is that the present invention sharpens design sketch to the original image of Fig. 3.
Fig. 5 is that prior art sharpens design sketch to the original image of Fig. 3.
Specific embodiment
Further illustrate technical scheme below in conjunction with the accompanying drawings and by specific embodiment.
Fig. 1 is the image sharpening method flow chart that the specific embodiment of the invention is provided.
A kind of image sharpening method, including：
Eight consecutive points around step a, selection decision-point, the decision-point and consecutive points are all the pixels in image
Point；
Step b, the consecutive points are calculated respectively with the GTG difference of the decision-point；
Step c, by the GTG difference with it is default sharpening decision threshold make comparisons, will be greater than the sharpening decision threshold
The number of the corresponding consecutive points of GTG difference add up as decision content；
Step d, judge whether the decision-point is marginal point according to decision content, if so, then the decision-point carried out sharp
Change is processed.
In the present embodiment, described image is made up of pixel one by one, and the edge contour of image is then by edge
Pixel namely group of edge points into, marginal point belong in the picture the color component of the high-frequency region of image, i.e. marginal point with
The color component change that the color component of its connected a direction area pixel point has violent change, marginal point will reach
Human eye can be allowed to see the color change of image, could be it is thought that the effective marginal point of image, by actual substantial amounts of test, people
Eye color change it is visible when, simultaneously changing value is different from multiple bases on single primary colours for pixel, substantially it is considered that single
Color change value is the twice of polychrome changing value, and in 256 GTG, marginal point monochrome changing value is about 40, marginal point during 64 GTG
Monochromatic changing value is about between 10 to 20.
In the present embodiment, the decision-point is sharpened specially：By the color component of the decision-point add or
Deduct the sharpening decision threshold.
In the present embodiment, it is described to judge whether the decision-point is that marginal point is specially according to decision content：Judge described
Whether within default value range, if so, then the decision-point is marginal point to decision content.
In the present embodiment, the GTG difference includes single color GTG difference and trichroism GTG difference, and the sharpening is sentenced
Threshold value is determined including first threshold and Second Threshold, the decision content includes the first decision content and the second decision content.
In the present embodiment, when the GTG difference is single color GTG difference, step c, step d are specially：
Step c1, the single color GTG difference is made comparisons with default first threshold, will be greater than the first threshold
The number of the corresponding consecutive points of single color GTG difference add up as the first decision content；
Step d1, judge first decision content whether within default value range, if so, then the decision-point is side
Edge point.
In the present embodiment, when the GTG difference is trichroism GTG difference, step c, step d are specially：
Step c2, the trichroism GTG difference is made comparisons with default Second Threshold, by trichroism GTG difference at least
Dichromatism GTG difference adds up to sentence for second more than the number of the corresponding consecutive points of trichroism GTG difference of the Second Threshold
Definite value；
Step d2, judge second decision content whether within default value range, if so, then the decision-point is side
Edge point.
In the present embodiment, the decision-point is as shown in Figure 2 with the position of the consecutive points.
In the present embodiment, in step a, the number of consecutive points is eight, and eight consecutive points are sentenced respectively positioned at described
The upper and lower, left and right of fixed point, upper left, upper right, lower-left and the direction of bottom right eight.
In the present embodiment, the default value range is more than or equal to two less than eight, when the decision content of the decision-point
During less than two, the decision-point is non-edge point；When the decision content of the decision-point is equal to eight, the decision-point is isolated
Noise spot.
In the present embodiment, marginal point around it close at least two points of eight points and decision-point grey jump
Value is more than default sharpening decision threshold, and the sharpening decision threshold includes the first threshold and Second Threshold.
In the present embodiment, also include after step d：
If step e, the decision-point are not marginal points, whether it is judged as isolated noise spot, if so, then to noise spot
Do smoothing processing, will noise spot color component be entered as the grey decision-making of the noise spot and the sharpening decision threshold and
Or it is poor.
By taking the original image shown in Fig. 3 as an example, Fig. 4 is that the present invention sharpens design sketch to the original image of Fig. 3；Fig. 5 is existing
There is technology to sharpen design sketch to the original image of Fig. 3.
In the present embodiment, image sharpening method of the invention is mainly used in driver IC, it is only necessary to take nine pixels
Plus-minus and the shift operation of GTG are done, the present invention only needs to do a step GTG difference subtraction, at most circulate two steps of four times
Compare a step additive operation, and step displacement and additive operation, the algorithm realizes simple, occupancy in driver IC designs
Logical resource it is few, the area Save power consumption for driver IC is fairly obvious, and the marginal point in the image for finding is more
And the filtration result of noise spot is remarkably reinforced.
The specific embodiment of the present invention is the foregoing is only, these descriptions are intended merely to explain the principle of the present invention, and
Can not be with any interpretation of structure as limiting the scope of the invention.Based on explanation herein, those skilled in the art is not
Other specific implementation methods that the present invention is associated by paying performing creative labour, these structures are needed to fall within the present invention
Protection domain within.
Claims (4)
1. a kind of image sharpening method, it is characterised in that include：
Eight consecutive points around step a, selection decision-point, the decision-point and consecutive points are all the pixels in image；
Step b, the consecutive points are calculated respectively with the GTG difference of the decision-point；
Step c, by the GTG difference with it is default sharpening decision threshold make comparisons, will be greater than it is described sharpening decision threshold ash
The number of the corresponding consecutive points of jump value adds up as decision content；
Step d, judge whether the decision-point is marginal point according to decision content, if so, then place is sharpened to the decision-point
Reason；
Wherein, in step a, the number of consecutive points is eight, eight consecutive points respectively positioned at the decision-point it is upper and lower,
Left and right, upper left, upper right, lower-left and the direction of bottom right eight；
It is described to judge whether the decision-point is that marginal point is specially according to decision content：Judge the decision content whether default
Within value range, if so, then the decision-point is marginal point；
The GTG difference includes single color GTG difference and trichroism GTG difference, and the sharpening decision threshold includes first threshold
And Second Threshold, the decision content includes the first decision content and the second decision content；
When the GTG difference is trichroism GTG difference, step c, step d are specially：
Step c2, the trichroism GTG difference is made comparisons with default Second Threshold, by least dichromatism in trichroism GTG difference
GTG difference adds up as the second decision content more than the number of the corresponding consecutive points of trichroism GTG difference of the Second Threshold；
Step d2, judge second decision content whether within default value range, if so, then the decision-point is edge
Point.
2. a kind of image sharpening method according to claim 1, it is characterised in that process is sharpened to the decision-point
Specially：By the color component of the decision-point plus or minus the sharpening decision threshold.
3. a kind of image sharpening method according to claim 1, it is characterised in that the default value range be more than etc.
Eight are less than in two, when the decision content of the decision-point is less than two, the decision-point is non-edge point；When sentencing for the decision-point
When definite value is equal to eight, the decision-point is isolated noise spot.
4. a kind of image sharpening method according to claim 3, it is characterised in that also include after step d：
If step e, the decision-point are not marginal points, whether it is judged as isolated noise spot, if so, then flat is done to noise spot
It is sliding to process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410025563.6A CN103763460B (en) | 2014-01-20 | 2014-01-20 | Image sharpening method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410025563.6A CN103763460B (en) | 2014-01-20 | 2014-01-20 | Image sharpening method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103763460A CN103763460A (en) | 2014-04-30 |
| CN103763460B true CN103763460B (en) | 2017-04-12 |
Family
ID=50530614
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410025563.6A Active CN103763460B (en) | 2014-01-20 | 2014-01-20 | Image sharpening method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103763460B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446908A (en) * | 2016-08-31 | 2017-02-22 | 乐视控股（北京）有限公司 | Method and device for detecting object in image |
| CN107742280A (en) | 2017-11-02 | 2018-02-27 | 浙江大华技术股份有限公司 | A kind of image sharpening method and device |
| CN107992182B (en) | 2017-12-05 | 2021-06-29 | 北京小米移动软件有限公司 | Method and device for displaying interface image |
| CN108989793A (en) * | 2018-07-20 | 2018-12-11 | 深圳市华星光电技术有限公司 | A kind of detection method and detection device of text pixel |
| CN112862851B (en) * | 2021-01-18 | 2021-10-15 | 网娱互动科技(北京)股份有限公司 | Automatic image matting method and system based on image recognition technology |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101742338A (en) * | 2008-11-05 | 2010-06-16 | 美格纳半导体有限会社 | Sharpness enhancing apparatus and method |
| CN102005051A (en) * | 2010-11-23 | 2011-04-06 | 华亚微电子（上海）有限公司 | Edge detection method and related device |
| CN102132554A (en) * | 2008-06-20 | 2011-07-20 | 惠普开发有限公司 | Method and system for efficient video processing |
-
2014
- 2014-01-20 CN CN201410025563.6A patent/CN103763460B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102132554A (en) * | 2008-06-20 | 2011-07-20 | 惠普开发有限公司 | Method and system for efficient video processing |
| CN101742338A (en) * | 2008-11-05 | 2010-06-16 | 美格纳半导体有限会社 | Sharpness enhancing apparatus and method |
| CN102005051A (en) * | 2010-11-23 | 2011-04-06 | 华亚微电子（上海）有限公司 | Edge detection method and related device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103763460A (en) | 2014-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103763460B (en) | Image sharpening method | |
| CN101783012B (en) | Automatic image defogging method based on dark primary colour | |
| CN101510302B (en) | Method and apparatus for enhancing image | |
| CN105488793B (en) | Image display method and image processing method | |
| CN108038833B (en) | Image self-adaptive sharpening method for gradient correlation detection and storage medium | |
| CN108022233A (en) | A kind of edge of work extracting method based on modified Canny operators | |
| CN105913396A (en) | Noise estimation-based image edge preservation mixed de-noising method | |
| CN105913385B (en) | Clarification method, system and the image processing apparatus of haze image | |
| CN101783861B (en) | Method and device for beautifying picture | |
| CN104657941B (en) | A kind of image border self-adapting enhancement method and device | |
| US8248662B2 (en) | Image forming apparatus and method thereof | |
| CN102306378B (en) | Image enhancement method | |
| CN109377450A (en) | A kind of edge-protected denoising method | |
| CN105117705A (en) | Iris image quality cascade type evaluation method | |
| CN104574327A (en) | Mammary gland image processing method and device | |
| CN106650728A (en) | Shadow license plate image binarization method | |
| CN103778611A (en) | Switch weighting vector median filter method utilizing edge detection | |
| CN104751422A (en) | Image enhancement and binaryzation method for photoprint file | |
| CN105118036A (en) | Method and system for restoring damaged image | |
| CN105243647A (en) | Linear spatial filtering-based image enhancement method | |
| CN114862843A (en) | Metal strip surface defect rapid detection method based on filter fusion | |
| CN107292898A (en) | A kind of car plate shadow Detection and minimizing technology based on HSV | |
| CN106780535A (en) | A kind of gray level image processing method | |
| CN105913391A (en) | Defogging method based on shape variable morphological reconstruction | |
| CN104299195A (en) | Self-adaptive dual-threshold and dark channel prior image defogging method based on |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20171127 Address after: 610015 Sichuan city of Chengdu province Tianfu Huayang Street North Street 40 No. 69 1 floor Patentee after: Chengdu core prestige Technology Co.,Ltd. Address before: 518102 A1, No. 5175, No. 5175, No. 310, 311, Yiyuan Road, Shenzhen, Shenzhen, Guangdong Province, 310, 311 Patentee before: SHENZHEN AIXIESHENG TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190117 Address after: 518102 District D and E, 7th Floor, Building 3, Tingwei Industrial Park, 6 Liufang Road, Xin'an Street, Baoan District, Shenzhen City, Guangdong Province Patentee after: SHENZHEN AIXIESHENG TECHNOLOGY Co.,Ltd. Address before: 610015 Floor 1, 69 Zhengbei Middle Street, Huayang Street, Tianfu New District, Chengdu City, Sichuan Province Patentee before: Chengdu core prestige Technology Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 518102 District D and E, 7th Floor, Building 3, Tingwei Industrial Park, 6 Liufang Road, Xin'an Street, Baoan District, Shenzhen City, Guangdong Province Patentee after: Shenzhen Aixiesheng Technology Co.,Ltd. Address before: 518102 District D and E, 7th Floor, Building 3, Tingwei Industrial Park, 6 Liufang Road, Xin'an Street, Baoan District, Shenzhen City, Guangdong Province Patentee before: SHENZHEN AIXIESHENG TECHNOLOGY Co.,Ltd. |