CN101706948B - Image amplifying method based on plum-blossom interpolation - Google Patents
Image amplifying method based on plum-blossom interpolation Download PDFInfo
- Publication number
- CN101706948B CN101706948B CN2009101941410A CN200910194141A CN101706948B CN 101706948 B CN101706948 B CN 101706948B CN 2009101941410 A CN2009101941410 A CN 2009101941410A CN 200910194141 A CN200910194141 A CN 200910194141A CN 101706948 B CN101706948 B CN 101706948B
- Authority
- CN
- China
- Prior art keywords
- interpolation
- image
- window
- points
- point
- 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
- 238000000034 method Methods 0.000 title claims abstract description 44
- 230000009545 invasion Effects 0.000 claims abstract description 9
- 230000008901 benefit Effects 0.000 claims description 7
- 238000013507 mapping Methods 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 230000003321 amplification Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 238000003199 nucleic acid amplification method Methods 0.000 description 4
- 239000000047 product Substances 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 239000012467 final product Substances 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 108091008695 photoreceptors Proteins 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000004304 visual acuity Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Landscapes
- Image Processing (AREA)
- Editing Of Facsimile Originals (AREA)
Abstract
The invention discloses an image amplifying method based on plum-blossom interpolation. The method comprises the following steps: 1) mapping an original image I to an image J to be interpolated so as to form a benchmark, namely J(L*i-1, L*j-1) is equal to I(i, j); 2) detecting the image, taking a window being 2*2 as a unit to form a grid, judging variance of the grid, if the variance is smaller than a set threshold, forming a smooth area and using a bilinear interpolation method, if not, forming an edge area and implementing the step 3); 3) for a middle point J of the window, calculating the value of the middle point by taking four benchmarks in the window based on the principle of strong invasion, and interpolating the middle point according to the calculated value; and 4) interpolating edge points including remaining points in the window after all the middle points are interpolated, utilizing two original benchmarks and two middle points nearby to form another window being 2*2, and repeating the step 2) for calculation. The method can lead edge or texture areas to keep clear while fast amplifying the images.
Description
Technical field
The invention belongs to image processing field, particularly relate to a kind of image magnification method based on plum-blossom interpolation.
Background technology
In a large amount of electronic image applications, people often expectation obtain high resolving power (being called for short HR) image, and high resolving power means the picture element density height in the image, and more details can be provided, and still can keep clear after image is exaggerated.If high-resolution image can be provided, the performance of the pattern-recognition in the computer vision will improve greatly.For present digital imaging system, the spatial resolution major decision of CCD/CMOS image photoreceptor the resolving limit of digital picture.Along with universal day by day with Flame Image Process of steadily improving of living standard, scientific research and practical application are more and more higher to the requirement of picture quality.But the level of resolution of these photoreceptors can not satisfy the needs of scientific research usually.For example, usually need to be similar to simulation 35 mm film, image do not have ghost image when amplifying ultrahigh resolution level in the scientific research process.
The target that image amplifies is devoted to keep minutia, promptly keeps edge feature.There has been certain methods in prior art, one class is based on the method for reconstruction, it is considered from the angle of image sampling, think that it is the inverse process of image down sampling that image amplifies, and therefore is transformed into inverse problem with the image amplification and finds the solution, and then customize various prior models according to edge or texture, such as the moment characteristics model, the figure at edge cuts model, and then uses maximum a posteriori algorithm for estimating (MAP) to find the solution enlarged image, has obtained effect preferably.
Second class is based on EVOLUTION EQUATION, use the method for partial differential equation (PDE), can keep edge feature by the gradient of control image border as anisotropic diffusion equation, or can remove sawtooth effect by the equi intensity curve (isophote) of control chart picture, but these class methods can produce the halation of some virtualizations at the smooth domain of image, and, make these class methods be absorbed in the bottleneck of application because partial differential equation is difficult to determine the intrinsic weakness of termination of iterations condition.
More than the shortcoming of two kinds of method maximums realize that exactly speed is very slow, be difficult to be applied to reality.The polynomial interpolation method that our modal improvement is traditional, such as partial gradient feature based on the edge, it has designed antigradient power, and in conjunction with traditional bilinear interpolation, bicubic interpolation algorithm, improved amplification effect to a certain extent, and kept the fireballing advantage of traditional interpolation method, but this class methods majority simply extends to two dimension with one-dimension method with the form of tensor product, because it is inseparable that image comes down to two dimension, so the blurring effect of edge or texture region is still very obvious.
Summary of the invention
The present invention is directed to above deficiency, proposed a kind of adaptive image magnification method, can make edge or texture region keep clear in the enlarged image fast based on plum-blossom interpolation.
In order to realize goal of the invention, the technical scheme of employing is:
A kind of image magnification method based on plum-blossom interpolation may further comprise the steps:
1) original image I is mapped to image J to be inserted, forms reference point, promptly J (L*i-1, L*j-1)=I (i, j), wherein L is an enlargement factor;
2) original image I being detected, is that a unit forms grid with the 2*2 window, judges the variance of grid, if variance less than setting threshold, then is a smooth region, then uses bilinear interpolation method, otherwise is fringe region, implementation step 3);
3) to the intermediate point of window, ask for the value of intermediate point with four reference points in the window based on the principle of advantage invasion, by the value interpolation intermediate point of being asked for, it is as follows specifically to ask for step:
A) at first four reference points are pressed gray-scale value rank order from big to small and formed a, b, c, four sequence of points of d;
B) calculate preceding two some a, b and latter two c, the gray-scale value between the d poor, with the gray scale difference value between a d and the some c with put b and put gray scale difference value between a as the interpolation weight factor, t is the gray-scale value of intermediate point,
4) interpolation edge point, comprise the intact all intermediate points of interpolation after, for remaining point in the window, utilize two original reference points and near two intermediate points to form the window of another 2*2, repeating said steps 2) ask for.
The present invention has following beneficial effect:
The method still is based upon on the basis of traditional polynomial interpolation method; therefore can realize zoom function apace; but the present invention has adopted a kind of interpolation method of blossom type; it is not the form of tensor product; be consistent with the two-dimentional inseparable essence of image; therefore the present invention can catch the acuteness feature at edge; the invention allows for a kind of surging invasion algorithm; the gray-scale value of interpolation point is leaned on toward surging district, thereby protected the acuteness of the edge of image after amplifying.
The present invention is applicable to various images, comprises gray-scale map and cromogram, but will carry out simple edge pre-detection before amplifying, and only edge-protected mechanism is implemented in the edge of image zone, and smooth region is then adopted general bilinear interpolation method.This kind interpolation strategies adaptively not only protected the edge of image feature, and further quickened algorithm, and be more effective in actual applications.
Description of drawings
In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, to do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the image magnification method process flow diagram that the present invention is based on plum-blossom interpolation;
Fig. 2 acts on the synoptic diagram at edge for the surging invasion of the present invention algorithm;
Fig. 3 is a plum-blossom interpolation each point synoptic diagram of the present invention;
Fig. 4 is the image amplification effect comparison diagram of the present invention and prior art.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making all other embodiment that obtained under the creative work prerequisite.
The present invention proposes a kind of adaptive image magnification method, can make edge or texture region keep clear in the enlarged image fast based on plum-blossom interpolation.
Below in conjunction with accompanying drawing the inventive method is further set forth.
The step of this method comprises:
1) mapping is former a bit to enlarged image: the point of original low-resolution image is mapped to enlarged image, form reference point;
2) judge level and smooth district and uneven skating area:
Treating the illustration picture, get four reference points and form a grid, at each grid, whether judge its variance less than certain setting threshold, if less than threshold value, then is level and smooth district, uses bilinear interpolation to get final product; If greater than threshold value, then be uneven skating area, then carry out subsequent step;
3) interpolation intermediate point:
At the intermediate point in each grid, adopt a kind of mode of advantageous point invasion to estimate its value, thereby can reach protection edge acuteness feature;
4) interpolation edge point: behind intact all intermediate points to be inserted, utilize original two reference points and two intermediate points to form a grid again again, go in the same way to handle at the intermediate point (being the edge point of image to be inserted) of grid.
Be illustrated in figure 1 as image magnification method process flow diagram based on plum-blossom interpolation.
If for the image I of the capable N row of a width of cloth M, enlargement factor is L=2
kDoubly, need enlarged image I, then carry out following steps:
Step 101, mapping form reference point: original image I is mapped to image J to be inserted, forms reference point, promptly J (L*i-1, L*j-1)=I (i, j);
Step 102, judge whether level and smooth district:
In order to accelerate the speed that image amplifies, in advance image is detected, be a unit with the 2*2 window, judge its variance, concrete computing formula is:
Var (I (i-1, j-1), I (i-1, j), I (i, j-1), I (i, j) }<th (default value of threshold value th generally can be 8).
If variance less than threshold value th, then is a smooth region, enter step 103, otherwise just explanation may be the region, edge, enters step 104;
Step 103, obtain point to be inserted with the bilinearity algorithm;
Variance is a smooth region less than threshold value th, then uses the bilinear interpolation technology to get final product, and promptly obtains point to be inserted with the bilinearity algorithm, and concrete formula is:
J(L*i-1,L*j-1)=(1-m)((1-n)I(i-1,j-1)+nI(i-1,j))+m((1-n)I(i,j-1)+nI(i,j))
Step 104, with advantage invasion algorithm interpolation intermediate point:
As shown in Figure 3, to the intermediate point (soft dot) of window, ask for based on the principle of advantage invasion with four reference points in the window.It is as follows specifically to ask for step:
A) at first four reference points are pressed gray-scale value rank order from big to small and formed a, b, c, four sequence of points of d;
B) calculate preceding two some a, b and latter two c, the gray-scale value between the d poor, as the interpolation weight factor, the more little then power of difference is big more with difference, illustrate that wait to ask a little should close 2 more little points of difference, gives prominence to advantage side with this;
As shown in Figure 2, for the edge, the gray-scale value of interpolation point leans on toward advantage side, and promptly t has promoted the result points t ' of bilinear interpolation, and the border width of the image after therefore amplifying can not become big along with amplification.
Step 105, interpolation edge point form edge point type grid:
As shown in Figure 3, interpolation for remaining point (grey round dot) in the window, utilizes two original reference points and near two intermediate points to form the window of another 2*2 after finishing all intermediate points, and repeating step 102 is asked for.
As shown in Figure 4, be the synoptic diagram that a butterfly (butterfly) image is amplified 2 times effect comparison.
Among Fig. 4, upper left figure is the existing design sketch that adopts nearest neighbor algorithm to obtain, the figure of lower left is the existing design sketch that adopts the bicubic algorithm to obtain, and top-right figure is the existing design sketch that adopts the bilinearity algorithm to obtain, and bottom-right figure is the design sketch that adopts the inventive method to obtain.Experiment shows that method effect of the present invention can be protected the edge effectively.Because algorithm of the present invention is linear calculating, so the efficient height, speed is fast, is easier to use in actual applications.
In sum, the present invention has following beneficial effect:
The method still is based upon on the basis of traditional polynomial interpolation method; therefore can realize zoom function apace; but the present invention has adopted a kind of interpolation method of blossom type; it is not the form of tensor product; be consistent with the two-dimentional inseparable essence of image; therefore the present invention can catch the acuteness feature at edge; the invention allows for a kind of surging invasion algorithm; the gray-scale value of interpolation point is leaned on toward surging district, thereby protected the acuteness of the edge of image after amplifying.
The present invention is applicable to various images, comprises gray-scale map and cromogram, but will carry out simple edge pre-detection before amplifying, and only edge-protected mechanism is implemented in the edge of image zone, and smooth region is then adopted general bilinear interpolation method.This kind interpolation strategies adaptively not only protected the edge of image feature, and further quickened algorithm, and be more effective in actual applications.
One of ordinary skill in the art will appreciate that all or part of step in the whole bag of tricks of the foregoing description is to instruct relevant hardware to finish by program, this program can be stored in the computer-readable recording medium, storage medium can comprise: ROM (read-only memory) (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), disk or CD etc.
More than to a kind of image magnification method that the embodiment of the invention provided based on plum-blossom interpolation, be described in detail, used specific case herein principle of the present invention and embodiment are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that all can change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.
Claims (1)
1. the image magnification method based on plum-blossom interpolation is characterized in that, may further comprise the steps:
1) original image I is mapped to image J to be inserted, forms reference point, promptly J (L*i-1, L*j-1)=I (i, j), wherein L is an enlargement factor;
2) original image I being detected, is that a unit forms grid with the 2*2 window, judges the variance of grid, if variance less than setting threshold, then is a smooth region, uses bilinear interpolation method, otherwise is fringe region, implementation step 3);
3) to the intermediate point of window, ask for the value of intermediate point with four reference points in the window based on the principle of advantage invasion, by the value interpolation intermediate point of being asked for, it is as follows specifically to ask for step:
A) at first four reference points are pressed gray-scale value rank order from big to small and formed a, b, c, four sequence of points of d;
B) calculate preceding two some a, b and latter two c, the gray-scale value between the d poor, with the gray scale difference value between a d and the some c with put b and put gray scale difference value between a as the interpolation weight factor, t is the gray-scale value of intermediate point,
4) interpolation edge point, comprise the intact all intermediate points of interpolation after, for remaining point in the window, utilize two original reference points and near two intermediate points to form the window of another 2*2, repeating said steps 2) ask for.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009101941410A CN101706948B (en) | 2009-11-26 | 2009-11-26 | Image amplifying method based on plum-blossom interpolation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009101941410A CN101706948B (en) | 2009-11-26 | 2009-11-26 | Image amplifying method based on plum-blossom interpolation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101706948A CN101706948A (en) | 2010-05-12 |
CN101706948B true CN101706948B (en) | 2011-11-30 |
Family
ID=42377170
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2009101941410A Expired - Fee Related CN101706948B (en) | 2009-11-26 | 2009-11-26 | Image amplifying method based on plum-blossom interpolation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101706948B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103034976B (en) * | 2011-09-29 | 2015-10-14 | 华晶科技股份有限公司 | The disposal route of image interpolarting |
CN103456031B (en) * | 2013-07-08 | 2016-03-30 | 江南大学 | A kind of new method of area image interpolation |
CN108596835B (en) * | 2018-05-04 | 2022-04-26 | 嘉兴南湖学院 | Image amplification method for circular area |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1667650A (en) * | 2005-04-08 | 2005-09-14 | 杭州国芯科技有限公司 | Image zooming method based on edge detection |
CN101216935A (en) * | 2008-01-17 | 2008-07-09 | 四川虹微技术有限公司 | Image amplification method based on spline function interpolation algorithm |
CN101499164A (en) * | 2009-02-27 | 2009-08-05 | 西安交通大学 | Image interpolation reconstruction method based on single low-resolution image |
-
2009
- 2009-11-26 CN CN2009101941410A patent/CN101706948B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1667650A (en) * | 2005-04-08 | 2005-09-14 | 杭州国芯科技有限公司 | Image zooming method based on edge detection |
CN101216935A (en) * | 2008-01-17 | 2008-07-09 | 四川虹微技术有限公司 | Image amplification method based on spline function interpolation algorithm |
CN101499164A (en) * | 2009-02-27 | 2009-08-05 | 西安交通大学 | Image interpolation reconstruction method based on single low-resolution image |
Non-Patent Citations (1)
Title |
---|
刘晓松等.基于统计特征的彩色图像快速插值方法.《电子学报》.2004,第32卷(第1期),29-33. * |
Also Published As
Publication number | Publication date |
---|---|
CN101706948A (en) | 2010-05-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105654438A (en) | Gray scale image fitting enhancement method based on local histogram equalization | |
Jing et al. | Fast image interpolation using directional inverse distance weighting for real-time applications | |
CN104217416A (en) | Grayscale image processing method and device | |
CN102306377A (en) | Method and device for reducing noise in ultrasound image | |
CN113221925B (en) | Target detection method and device based on multi-scale image | |
CN106169173B (en) | Image interpolation method | |
CN101706948B (en) | Image amplifying method based on plum-blossom interpolation | |
CN102521813A (en) | Infrared image adaptive enhancement method based on dual-platform histogram | |
Yang et al. | An improved iterative back projection algorithm based on ringing artifacts suppression | |
CN101976336A (en) | Fuzzy enhancement and surface fitting-based image edge characteristic extraction method | |
Mahajan et al. | Adaptive and non-adaptive image interpolation techniques | |
Liao et al. | Efficient image dehazing using boundary conditions and local contrast | |
CN102096913B (en) | Multi-strategy image fusion method under compressed sensing framework | |
Zhou et al. | Optimal bi-directional seam carving for compressibility-aware image retargeting | |
CN106342330B (en) | A kind of image enchancing method of the gamma correction based on infrared image | |
CN114820401A (en) | Method for enhancing marine backlight infrared image by combining histogram transformation and edge information | |
CN102800047B (en) | Method for reconstructing super resolution of single-frame image | |
Kang et al. | Fusing soft-decision-adaptive and bicubic methods for image interpolation | |
Singh et al. | Single-image super-resolution using orthogonal rotation invariant moments | |
Shen et al. | A MAP algorithm to super-resolution image reconstruction | |
Wang et al. | A fast algorithm for three-dimensional otsu’s thresholding method | |
CN100538740C (en) | Method for lowering noise of medical sonogram anisotropic diffusion | |
CN106611405B (en) | Image interpolation method and device | |
Lin et al. | Dilated residual shrinkage network for SAR image despeckling | |
CN110136061B (en) | Resolution improving method and system based on depth convolution prediction and interpolation |
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 | ||
DD01 | Delivery of document by public notice |
Addressee: Chen Lina Document name: payment instructions |
|
DD01 | Delivery of document by public notice | ||
DD01 | Delivery of document by public notice |
Addressee: Guangdong Guanglian Electronic Technology Co.,Ltd. Person in charge of patents Document name: Notice of Termination of Patent Rights |
|
DD01 | Delivery of document by public notice | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20111130 |
|
CF01 | Termination of patent right due to non-payment of annual fee |