CN101188017A - Digital image zooming method and system - Google Patents
Digital image zooming method and system Download PDFInfo
- Publication number
- CN101188017A CN101188017A CNA2007101725022A CN200710172502A CN101188017A CN 101188017 A CN101188017 A CN 101188017A CN A2007101725022 A CNA2007101725022 A CN A2007101725022A CN 200710172502 A CN200710172502 A CN 200710172502A CN 101188017 A CN101188017 A CN 101188017A
- Authority
- CN
- China
- Prior art keywords
- point
- interpolation
- convergent
- divergent
- module
- 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
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention provides a zooming method and a zooming system of a data image, and includes an interpolation-point space position calculation module that is used for calculating the specific position of the prior interpolation point in an original image and calculating the correspondent zooming ratio, a correlation-degree calculation module that is used for calculating the correlation degree on different directions according to the different positions of the interpolation points, a classification judging module that is used for classifying the interpolation points on different directions, a reference-point and correspondent direction obtaining module that is used for selecting four reference points on the most relevant corresponding direction and calculating the corresponding direction of the interpolation point in the four reference points that have the most correlation degree, and an inner-interpolation module that is used for the interpolation on different directions. Under the condition that the integration is rather simpler, the method of the invention can select the appropriate correlation-point information for the interpolation according to the content of the image, and obtain relevant reference-point data so as to reform the interpolation point value according to the inner interpolation direction, thereby effectively protecting the border of the object and suppressing the vagueness of the valuable information.
Description
Technical field
The invention belongs to digital image processing field, specifically, relate to a kind of Zoom method and system of digital picture.
Background technology
In flat panel TV,, need main drive controlling chip SCALER to carry out the space convergent-divergent because input picture is different with lines per picture with the every capable pixel of the output image that needs to show.The space convergent-divergent of so-called image is an image interpolation, and is general when image amplifies, the high fdrequency component that can lose image, and it is fuzzyyer that image can seem, and complex image Zoom method effect is good more, but realizes but very difficult on integrated circuit.
Disclosed a large amount of image zoom algorithms in the prior art, for example, application number is to disclose a kind of digital image scaling method of designing integrated circuit in 03111801.1 the Chinese patent, in described method upwards convergent-divergent and downwards convergent-divergent realize by single circuit; Application number is the method that 200410102692.7 Chinese patent also discloses a kind of image zoom, and described method is carried out convergent-divergent by setting up look-up table, has reduced the complexity of algorithm; In addition, application number is also to disclose a kind of adaptive space Zoom method that is used for interlaced video signal in 02817340.6 the Chinese patent, wherein propose sports ground based on the field convergent-divergent, and frozen frozen mass is based on the frame convergent-divergent.But, more than each method a common shortcoming is all arranged, be exactly not consider the edge of image direction, in interpolation process, only utilized spatial positional information to judge interpolation point and on every side correlativity, interpolation coefficient is determined in advance, has nothing to do with picture material.Like this at object boundary or details place, incoherent two parts content generation interpolation point that is combined fuzzyly just is difficult to avoid, and therefore edge of image partly can be blured in above each existing method.
Summary of the invention
The object of the present invention is to provide a kind of Zoom method and system of digital picture, it can protect object boundary effectively, thereby suppresses the obfuscation of image border useful information, and is easy to the realization of integrated circuit.
For achieving the above object, the invention provides a kind of Zoom method of digital picture, it comprises following steps: calculate current particular location in original image, and calculate corresponding scaling; According to interpolation point position difference, calculate the degree of correlation of all directions; Point to different directions is sorted out; Choose maximally related counterparty to four neighborhood points; Calculate the counterparty of interpolation point in maximally related four reference point of choosing to; Carry out interpolation along direction.
The Zoom method of digital picture of the present invention, the formula that wherein calculates corresponding scaling is:
x＝floor(i
n*Height/H
n+1)
y＝floor(j
n*Width/W
n+1)
Δx＝i
n*Height/H
n-x+1；
Δy＝j
n*Width/W
n-y+1；
I wherein
n, j
nBe the position of the pixel of the new images behind the convergent-divergent, x, y are the position of picture point before the pairing convergent-divergent of this picture element, and Δ x, Δ y are point behind this convergent-divergent 4 nearest distance in the image before the convergent-divergent, W
n, H
nWide and high for image behind the convergent-divergent, Width, Height are the wide and high of image before the convergent-divergent.
The Zoom method of digital picture of the present invention, the step of wherein calculating the degree of correlation of all directions is in more detail: the insertion point of zones of different is judged the degree of correlation at particular pixels point place respectively on 0 °, 45 °, 90 ° and 135 ° of four directions.
The Zoom method of digital picture of the present invention wherein along the step that direction is carried out interpolation is in more detail: adopt three continuous polynomial interpolation kernels of 2 * 2 neighborhoods, computing formula is:
I (X wherein
Ij) be X
IjThe brightness of place's sampled point, X
IjBe the summit of relevant neighborhood, h () is the interpolation kernel.
The Zoom method of digital picture of the present invention, input be the wide of convergent-divergent front court (frame) with the wide and high of high, convergent-divergent back court (frame) and when front court (frame) information, that will export at last is convergent-divergent back court (frame) information.
The present invention also provides a kind of panntographic system of digital picture, it comprises: interpolation point locus computing module, be used for calculating current particular location at original image, and calculate corresponding scaling, described interpolation point locus computing module be input as wide and high W when front court (frame) information, convergent-divergent back court (frame)
n, H
n, and wide and high Width, the Height of convergent-divergent front court (frame), be output as interpolation point spatial positional information Δ x, Δ y, picture element position x, the y of corresponding original image; The relatedness computation module is used for the degree of correlation according to position, insertion point various computing all directions, described relatedness computation module be input as Δ y, be output as the degree of correlation of all directions; The classification determination module is used for the point of different directions is sorted out, and the degree of correlation that is input as all directions of described classification determination module is output as the direction DFlag that classification is judged; Obtain reference point and counterparty to module, be used to choose maximally related counterparty to four neighborhood points, and calculate the counterparty of interpolation point in maximally related four reference point of choosing to, the described picture element position x, the y that are input as DFlag, corresponding original image, interpolation point spatial positional information Δ x, the Δ y that obtains reference point and counterparty to module, new direction Δ x ', Δ y ' after being output as four reference point locations and projecting to new coordinate; Interpose module is used for carrying out interpolation along direction, and new direction Δ x ', Δ y ' after being input as four reference point locations and projecting to new coordinate of described interpose module are output as field (frame) image behind the convergent-divergent.
Digital image scaling method of the present invention is not under the integrated circuit complexity is very high situation; can select suitable reference point information to carry out interpolation according to picture material; take out corresponding reference point data then; in conjunction with interior direction interpolation reconstruct interpolation point numerical value; obviously adopt digital image scaling method of the present invention can protect object boundary effectively, thereby suppress the fuzzy of image border useful information.
Description of drawings
To the description of a preferred embodiment of the present invention, can further understand purpose, specific structural features and the advantage of its invention by following in conjunction with its accompanying drawing.Wherein, accompanying drawing is:
Fig. 1 is the block diagram of a preferred embodiment of the panntographic system of digital picture of the present invention;
Fig. 2 is the process flow diagram of a preferred embodiment of the Zoom method of digital picture of the present invention;
Fig. 3 is locus, an insertion point synoptic diagram in the preferred embodiment of Zoom method of digital picture of the present invention;
Choose the counterparty in the preferred embodiment of Fig. 4 for the Zoom method of digital picture of the present invention and put synoptic diagram to neighborhood;
Fig. 5 is interpolation direction new in the preferred embodiment of Zoom method of digital picture of the present invention and the geometric relationship synoptic diagram between former direction.
Embodiment
Below in conjunction with Fig. 1 to Fig. 5, specify a better embodiment of the present invention.
Fig. 1 is the block diagram of a preferred embodiment of the panntographic system of digital picture of the present invention.As shown in the figure, the panntographic system 1 of digital picture of the present invention comprises: interpolation point locus computing module 10, relatedness computation module 12, classification determination module 14, obtain reference point and counterparty to module 16 and interpose module 18.
In more detail, this interpolation point locus computing module 10, be used for calculating current particular location, and calculate corresponding scaling at original image, described interpolation point locus computing module 10 be input as wide and high W when front court (frame) information, convergent-divergent back court (frame)
n, H
n, and wide and high Width, the Height of convergent-divergent front court (frame), be output as interpolation point spatial positional information Δ x, Δ y, picture element position x, the y of corresponding original image.This relatedness computation module 12 is used for the degree of correlation according to position, insertion point various computing all directions, described relatedness computation module 12 be input as Δ y, be output as the degree of correlation of all directions.This determination module 14 of classifying is used for the point of different directions is sorted out, and the degree of correlation that is input as all directions of described classification determination module 14 is output as the direction DFlag that classification is judged.This obtains reference point and counterparty to module 16, be used to choose maximally related counterparty to four neighborhood points, and calculate the counterparty of interpolation point in maximally related four reference point of choosing to, the described picture element position x, the y that are input as DFlag, corresponding original image, interpolation point spatial positional information Δ x, the Δ y that obtains reference point and counterparty to module 16, new direction Δ x ', Δ y ' after being output as four reference point locations and projecting to new coordinate.This interpose module 18 is used for carrying out interpolation along direction, and new direction Δ x ', Δ y ' after being input as four reference point locations and projecting to new coordinate of described interpose module are output as field (frame) image behind the convergent-divergent.
Then, see also Fig. 2, be the process flow diagram of a preferred embodiment of the Zoom method of digital picture of the present invention.
At first, execution in step S1, the scan line input, and obtain relevant information.In the present embodiment, described relevant information mainly comprises the wide and high W of (frame) information and convergent-divergent back court (frame)
n, H
n, wide and high Width, the Height of convergent-divergent front court (frame).
Then, execution in step S2, current particular location in original image is promptly calculated in the computer memory position, calculates corresponding scaling simultaneously.In the present embodiment, adopt following computing formula to calculate described locus:
x＝floor(i
n*Height/H
n+1)
y＝floor(j
n*Width/W
n+1)
Δx＝i
n*Height/H
n-x+1；
Δy＝j
n*Width/W
n-y+1；
I wherein
n, j
nBe the position of the pixel of the new images behind the convergent-divergent, x, y are the position of picture point before the pairing convergent-divergent of this picture element, and Δ x, Δ y are point behind this convergent-divergent 4 nearest distance in the image before the convergent-divergent, W
n, H
nWide and high for image behind the convergent-divergent, Width, Height are the wide and high of image before the convergent-divergent.
Then, execution in step S3 according to interpolation point position difference, calculates the degree of correlation of all directions.In the present embodiment, please consult Fig. 3 simultaneously, Δ y≤0.5 (as the white portion in the square frame) selects a
1, a
2, a
3, b
1, b
2, b
3Up and down two each three column data of row judge the related side to; A is then selected in Δ y＞0.5 (as the net region in the square frame)
2, a
3, a
4, b
2, b
3, b
4Up and down two each three column data of row judge the related side to, wherein, the point of zones of different is judged pixel (x, the correlativity of y) locating respectively on 0 °, 45 °, 90 ° and 135 ° of four directions.
Relevant direction determining is made in Δ y≤0.5 o'clock, and its formula is:
D
0＝|a
2-b
2|+|a
3-b
3|
D
1＝|a
2-b
1|+|a
3-b
2|
D
-1＝|a
1-b
2|+|a
2-b
3|
D
H＝(|a
3-a
1|+|a
3-a
2|+|b
3-b
1|+|b
3-b
2|)/2
D
min＝min(D
0，D
1，D
-1，D
H)
If Δ y＞0.5, the picture element that moves to right when doing direction determining, its formula is:
D
0＝|a
2-b
2|+|a
3-b
3|
D
1＝|a
4-b
3|+|a
3-b
2|
D
-1＝|a
2-b
3|+|a
3-b
4|
D
H＝(|a
4-a
2|+|a
4-a
3|+|b
4-b
2|+|b
4-b
3|)/2
D
min＝min(D
0，D
1，D
-1，D
H)
As shown in Figure 3, b
3Some D (i for target image
d, j
d) after converting in the space, corresponding to S (i before the image zoom
s, j
s) the point.D
0, D
1, D
-1, D
HNumerical values recited represent the make progress correlativity of pixel of interpolation and counterparty, it is strong more to be worth more little correlativity, carries out interpolation at the direction of correlativity maximum.Certainly, 10 or ten four can be carried out more multidirectional judgement, can realize to those skilled in the art, do not give unnecessary details again at this.
D wherein
Min=D
0Represent vertical direction, D
Min=D
1Represent 45 degree directions, D
Min=D
-1Represent 135 degree directions, D
Min=D
HRepresent horizontal direction, obviously so in the present embodiment judgement is also unreliable, therefore needs to eliminate unreliable judgement source, the additional information of following message for helping to judge:
D
01＝|a
2-b
2|
D
02＝|a
3-b
3|
if?Δy≤0.5
D
H1＝(|a
3-a
1|+|a
3-a
2|)/2
D
H2＝(|b
3-b
1|+|b
3-b
2|)/2
else
D
H1＝(|a
4-a
2|+|a
4-a
2|)/2
D
H2＝(|b
4-b
2|+|b
4-b
3|)/2
end
Then, execution in step S4 sorts out the point of different directions.
In the present embodiment,, need filter,, filter out the unreliable edge that noise or texture cause in the hope of obtaining real edge to detected edge because the edge is very sensitive to noise.The distinguishing feature at the edge of standard be exactly correlativity apparently higher than its vertical direction, yet its edge direction such as noise does not have dominance, the difference stochastic distribution of all directions also has less difference value in the direction perpendicular to minimal gradient usually.Such noise spot does not have clear and definite direction, selects for use near its four point to carry out interpolation, and it is the same disposing with the edge of vertical direction.
Point for 135 degree can be provided with and be masked as-1; Point for 45 degree can be provided with and be masked as 1; For the point of 0 degree, if two row differences many up and down illustrates that this is horizontal boundary,, need use nearest-neighbor method in order to prevent to blur, can be provided with and be masked as 2; For the point on the perpendicular line, if about two row differences many, illustrate that this is the boundary of perpendicular line, in order to prevent to blur, need use nearest-neighbor method, can be provided with and be masked as 3; Point or the uncertain point of direction for 90 degree can be provided with and be masked as 0; Can obtain following decision condition thus:
if(D
min＝＝D
-1&|D
-1-D
1|＞T
b&D
min＜T
s)
DFlag＝-1
elseif(D
min＝＝D
1&|D
-1-D
1|＞T
b&D
min＜T
s)
DFlag＝1
elseif(D
min＝＝D
H&|D
H1-D
H2|＞T
b&D
min＜T
s)
DFlag＝2
elseif(D
min＝＝D
0&|D
01-D
02|＞T
b&D
min＜T
s
DFlag＝3
else
DFlag＝0
end
Here T
bAnd T
sBe threshold value.Common optional T
b=30, T
s=75.
Then execution in step S5, choose maximally related counterparty to four reference point.In the present embodiment, according to the sign that step S4 is provided with, can find corresponding maximally related four reference point.Please in conjunction with shown in Figure 4, in two kinds of situation:
1. when Δ y≤0.5 (as the solid line point)
Current point be masked as DFlag=-1 the time, choose a
1, a
2, b
2, b
3These four points; Current point be masked as DFlag=1 the time, choose a
2, a
3, b
1, b
2These four points; Current point be masked as DFlag=0 the time, choose a
2, a
3, b
2, b
3These four points; Current point be masked as DFlag=2 the time, judge the size of Δ x, if a is chosen in Δ x≤0.5
2, a
3Mean value, if b is chosen in Δ x＞0.5
2, b
3Mean value; Current point be masked as DFlag=3 the time, judge the size of Δ y, if a is chosen in Δ y≤0.5
2, b
2Mean value, if a is chosen in Δ y＞0.5
3, b
3Mean value.
2. when Δ y＞0.5 (as the dotted line point)
Current point be masked as DFlag=-1 the time, choose a
2, a
3, b
3, b
4These four points; Current point be masked as DFlag=1 the time, choose a
3, a
4, b
2, b
3These four points; Current point be masked as DFlag=0 the time, choose a
2, a
3, b
2, b
3These four points; Current point be masked as DFlag=2 the time, judge the size of Δ x, if a is chosen in Δ x≤0.5
2, a
3Mean value, if b is chosen in Δ x＞0.5
2, b
3Mean value; Current point be masked as DFlag=3 the time, judge the size of Δ y, if a is chosen in Δ y≤0.5
2, b
2Mean value, if a is chosen in Δ y＞0.5
3, b
3Mean value.
Then, execution in step S6, calculate the counterparty of interpolation point in maximally related four reference point of choosing to.In the present embodiment, please in conjunction with shown in Figure 5, new interpolation direction and the geometric relationship between former direction.Former direction Δ x, Δ y are the numerical value of interpolation point relative coordinate x-y, and in relevant neighborhood quadrilateral, the direction that projects to new coordinate x '-y ' is Δ x ', Δ y '.
In the time of DFlag=-1 or DFlag=1, need project to tetragonal limit again.Find the solution through geometric relationship, recomputate new Δ x ', Δ y '; When DFlag=0, then use original Δ x, Δ y.When DFlag=2 or DFlag=3, then needn't calculate, as described in step S5, directly choose mean value near 2.
In the present embodiment, Δ x ', Δ y ' are calculated as follows:
Δx′＝Δx
Δy′＝Δy+(Δx-1)DFlag
At last, execution in step S7 carries out interpolation along direction.In the present embodiment, be three continuous polynomial interpolation kernels that adopt 2 * 2 neighborhoods, computing formula is:
I (X in the formula
Ij) be X
IjThe brightness of place's sampled point, X
IjBe the summit of relevant neighborhood, h () is the interpolation kernel.Though be limited in 2 * 2 zones because the spatial domain response curve of three continuous polynomial interpolation kernels is the same with bilinear interpolation, but it is mild than the bilinear interpolation border transition, the sidelobe of its frequency spectrum obviously reduces, and can improve the too fuzzy shortcoming of bilinear interpolation to a certain extent.
The method that the present invention proposes can select suitable reference point information to carry out interpolation according to picture material under the integrated circuit complexity is not very high situation, takes out corresponding reference point data in that this side up then, in conjunction with interior direction interpolation reconstruct interpolation point numerical value.Obviously the method can be protected object boundary effectively, suppresses the fuzzy of useful information.
In sum, the invention provides a kind of Zoom method of digital picture, its choice of location direction according to the insertion point is judged reference point, different edges adopts different disposal routes, can effectively solve the ill-defined shortcoming of traditional Zoom method, and the panntographic system structure of digital picture of the present invention is comparatively simple, so be fit to be applied on the integrated circuit.
Of particular note, the Zoom method of digital picture of the present invention is not limited to step execution sequence defined in the foregoing description, although the present invention is had been described in detail with reference to preferred embodiment, those of ordinary skill in the art is to be understood that, can make amendment or be equal to replacement the present invention, and not breaking away from the spirit and scope of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.
Claims (8)
1. the Zoom method of a digital picture is characterized in that, comprises following steps:
Calculate current particular location in original image, and calculate corresponding scaling;
According to interpolation point position difference, calculate the degree of correlation of all directions;
Point to different directions is sorted out;
Choose maximally related counterparty to four reference point;
Calculate the counterparty of interpolation point in maximally related four reference point of choosing to;
Carry out interpolation along direction.
2. the Zoom method of digital picture according to claim 1 is characterized in that, at first will import the wide and high of the wide of convergent-divergent front court (frame) and height, convergent-divergent back court (frame) and work as front court (frame) information.
3. the Zoom method of digital picture according to claim 2 is characterized in that, the formula that wherein calculates corresponding scaling is:
x＝floor(i
n*Height/H
n+1)
y＝floor(j
n*Width/W
n+1)
Δx＝i
n*Height/H
n-x+1；
Δy＝j
n*Width/W
n-y+1；
I wherein
n, j
nBe the position of the pixel of the new images behind the convergent-divergent, x, y are the position of picture point before the pairing convergent-divergent of this picture element, and Δ x, Δ y are point behind this convergent-divergent 4 nearest distance in the image before the convergent-divergent, W
n, H
nWide and high for image behind the convergent-divergent, Width, Height are the wide and high of image before the convergent-divergent.
4. the Zoom method of digital picture according to claim 1, it is characterized in that the step of wherein calculating the degree of correlation of all directions is in more detail: the insertion point of zones of different is judged the degree of correlation at particular pixels point place respectively on 0 °, 45 °, 90 ° and 135 ° of four directions.
5. the Zoom method of digital picture according to claim 1 is characterized in that, wherein the step of carrying out interpolation along direction is in more detail: three continuous polynomial interpolation kernels that adopt 2 * 2 neighborhoods.
6. the Zoom method of digital picture according to claim 5 is characterized in that, wherein adopts the formula of three continuous polynomial interpolation kernels of 2 * 2 neighborhoods to be:
I (X wherein
Ij) be X
IjThe brightness of place's sampled point, X
IjBe the summit of relevant neighborhood, h () is the interpolation kernel.
7. the Zoom method of digital picture according to claim 1 is characterized in that, output is convergent-divergent back court (frame) information at last.
8. the panntographic system of a digital picture is characterized in that, the panntographic system of described digital picture comprises:
Interpolation point locus computing module is used for calculating current particular location at original image, and calculates corresponding scaling, described interpolation point locus computing module be input as wide and high W when front court (frame) information, convergent-divergent back court (frame)
n, H
n, and wide and high Width, the Height of convergent-divergent front court (frame), be output as interpolation point spatial positional information Δ x, Δ y, picture element position x, the y of corresponding original image;
The relatedness computation module is used for the degree of correlation according to position, insertion point various computing all directions, described relatedness computation module be input as Δ y, be output as the degree of correlation of all directions;
The classification determination module is used for the point of different directions is sorted out, and the degree of correlation that is input as all directions of described classification determination module is output as the direction DFlag that classification is judged;
Obtain reference point and counterparty to module, be used to choose maximally related counterparty to four neighborhood points, and calculate the counterparty of interpolation point in maximally related four reference point of choosing to, the described picture element position x, the y that are input as DFlag, corresponding original image, interpolation point spatial positional information Δ x, the Δ y that obtains reference point and counterparty to module, new direction Δ x ', Δ y ' after being output as four reference point locations and projecting to new coordinate;
Interpose module is used for carrying out interpolation along direction, and new direction Δ x ', Δ y ' after being input as four reference point locations and projecting to new coordinate of described interpose module are output as field (frame) image behind the convergent-divergent.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007101725022A CN101188017A (en) | 2007-12-18 | 2007-12-18 | Digital image zooming method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007101725022A CN101188017A (en) | 2007-12-18 | 2007-12-18 | Digital image zooming method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101188017A true CN101188017A (en) | 2008-05-28 |
Family
ID=39480392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007101725022A Pending CN101188017A (en) | 2007-12-18 | 2007-12-18 | Digital image zooming method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101188017A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100596173C (en) * | 2008-06-11 | 2010-03-24 | 四川虹微技术有限公司 | Image zooming method and apparatus |
| CN102609900A (en) * | 2012-01-16 | 2012-07-25 | 北京环星基业数码科技发展有限公司 | Image interpolation algorithm based on region consistency |
| CN103093418A (en) * | 2013-02-21 | 2013-05-08 | 深圳市晶日盛科技有限公司 | Improved digital image resizing method |
| CN103714512A (en) * | 2013-12-25 | 2014-04-09 | 中国船舶重工集团公司第七0九研究所 | Real-time color image scaling method based on Lanczos core |
| CN104134189A (en) * | 2014-07-31 | 2014-11-05 | 青岛海信信芯科技有限公司 | Method and device for magnifying image |
| CN105096248A (en) * | 2014-05-09 | 2015-11-25 | 北大方正集团有限公司 | Image processing method and equipment |
| WO2017096814A1 (en) * | 2015-12-07 | 2017-06-15 | 乐视控股（北京）有限公司 | Image processing method and apparatus |
| CN107105210A (en) * | 2017-05-25 | 2017-08-29 | 歌尔股份有限公司 | Projection correction's method and device |
| CN109523469A (en) * | 2018-11-16 | 2019-03-26 | 深圳朗田亩半导体科技有限公司 | Image-scaling method and device |
| CN117726557A (en) * | 2023-05-17 | 2024-03-19 | 荣耀终端有限公司 | Image processing method, electronic device and storage medium |
-
2007
- 2007-12-18 CN CNA2007101725022A patent/CN101188017A/en active Pending
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100596173C (en) * | 2008-06-11 | 2010-03-24 | 四川虹微技术有限公司 | Image zooming method and apparatus |
| CN102609900A (en) * | 2012-01-16 | 2012-07-25 | 北京环星基业数码科技发展有限公司 | Image interpolation algorithm based on region consistency |
| CN103093418B (en) * | 2013-02-21 | 2015-08-26 | 深圳市晶日盛科技有限公司 | A kind of digital image scaling method of improvement |
| CN103093418A (en) * | 2013-02-21 | 2013-05-08 | 深圳市晶日盛科技有限公司 | Improved digital image resizing method |
| WO2014127665A1 (en) * | 2013-02-21 | 2014-08-28 | 深圳市晶日盛科技有限公司 | Improved method for digital image scaling |
| CN103714512B (en) * | 2013-12-25 | 2016-07-06 | 中国船舶重工集团公司第七0九研究所 | A kind of real-time color image-scaling method based on Lanczos core |
| CN103714512A (en) * | 2013-12-25 | 2014-04-09 | 中国船舶重工集团公司第七0九研究所 | Real-time color image scaling method based on Lanczos core |
| CN105096248A (en) * | 2014-05-09 | 2015-11-25 | 北大方正集团有限公司 | Image processing method and equipment |
| CN105096248B (en) * | 2014-05-09 | 2018-02-23 | 北大方正集团有限公司 | A kind of image processing method and equipment |
| CN104134189A (en) * | 2014-07-31 | 2014-11-05 | 青岛海信信芯科技有限公司 | Method and device for magnifying image |
| CN104134189B (en) * | 2014-07-31 | 2017-07-28 | 青岛海信电器股份有限公司 | A kind of method and device of image amplification |
| WO2017096814A1 (en) * | 2015-12-07 | 2017-06-15 | 乐视控股（北京）有限公司 | Image processing method and apparatus |
| CN107105210A (en) * | 2017-05-25 | 2017-08-29 | 歌尔股份有限公司 | Projection correction's method and device |
| CN109523469A (en) * | 2018-11-16 | 2019-03-26 | 深圳朗田亩半导体科技有限公司 | Image-scaling method and device |
| CN109523469B (en) * | 2018-11-16 | 2022-12-27 | 深圳朗田亩半导体科技有限公司 | Image scaling method and device |
| CN117726557A (en) * | 2023-05-17 | 2024-03-19 | 荣耀终端有限公司 | Image processing method, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101188017A (en) | Digital image zooming method and system | |
| CN112104854A (en) | Method and system for robust virtual view generation between camera views | |
| US7149355B2 (en) | Image processing apparatus, image processing method, image processing program, and computer-readable record medium storing image processing program | |
| US11151704B2 (en) | Apparatus and methods for artifact detection and removal using frame interpolation techniques | |
| US6252577B1 (en) | Efficient methodology for scaling and transferring images | |
| US20060244861A1 (en) | Method for detecting bisection pattern in deinterlacing | |
| US20080056617A1 (en) | Method and apparatus for interpolating image information | |
| CN106204441B (en) | Image local amplification method and device | |
| US7869665B2 (en) | Composite method and apparatus for adjusting image resolution | |
| US8711231B2 (en) | Digital image processing device and processing method thereof | |
| US7751642B1 (en) | Methods and devices for image processing, image capturing and image downscaling | |
| US20100026888A1 (en) | Image processing method and system with repetitive pattern detection | |
| WO2014008329A1 (en) | System and method to enhance and process a digital image | |
| JP2007249436A (en) | Image signal processor and processing method | |
| Cho et al. | Temporal incoherence-free video retargeting using foreground aware extrapolation | |
| US7532773B2 (en) | Directional interpolation method and device for increasing resolution of an image | |
| US20120008692A1 (en) | Image processing device and image processing method | |
| Lien et al. | Efficient VLSI architecture for edge-oriented demosaicking | |
| Gour et al. | Hardware accelerator for real-time image resizing | |
| JP2009094862A (en) | Video signal processor | |
| US7957594B2 (en) | Outline definition apparatus and outline definition method, and image processing apparatus | |
| JP4055908B2 (en) | Wipe pattern generator | |
| EP2071512B1 (en) | Method for generating distances representative of the edge orientations in a video picture, corresponding device and use of the method for deinterlacing or format conversion | |
| JP5377649B2 (en) | Image processing apparatus and video reproduction apparatus | |
| US20060120615A1 (en) | Frame compensation for moving imaging devices |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |