CN107909544A - A kind of method for correcting image and system - Google Patents
A kind of method for correcting image and system Download PDFInfo
- Publication number
- CN107909544A CN107909544A CN201711043165.7A CN201711043165A CN107909544A CN 107909544 A CN107909544 A CN 107909544A CN 201711043165 A CN201711043165 A CN 201711043165A CN 107909544 A CN107909544 A CN 107909544A
- Authority
- CN
- China
- Prior art keywords
- input picture
- pixel
- matrix
- boundary line
- original image
- 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
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/60—Rotation of a whole image or part thereof
- G06T3/608—Skewing or deskewing, e.g. by two-pass or three-pass rotation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4007—Interpolation-based scaling, e.g. bilinear interpolation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4023—Decimation- or insertion-based scaling, e.g. pixel or line decimation
Abstract
The invention discloses a kind of method for correcting image and system, this method includes：Analysis is carried out to original image to define boundaries and vertex, correction parameter is determined according to boundary line and vertex, then carries out positive correction or reverse correction to image according to correction parameter.So technique according to the invention scheme, terminal is after original image is uploaded, system directly can analyze master pattern, then original image is positioned and corrected, it is accurate and quick, it is more intelligent, therefore the equipment of Image Acquisition is no longer limited to scanner, can extend to more equipment such as mobile phone, digital camera.
Description
Technical field
This application involves technical field of image processing, more particularly to a kind of method for correcting image and system.
Background technology
With the continuous progress and development of science and technology, the application case for automatic identification of taking pictures is more and more extensive, such as：Silver
The applications such as the fixation and recognition of row card and identity card, plate location recognition, it is various applications that target area is quickly found out from photo
One of key issue.
" wisdom packaging " application system is to allow user to gather label image using scanner, and high in the clouds is uploaded to by network,
Server receives the image that user uploads beyond the clouds, and processing analysis is carried out to image using image procossing application program.Wherein, exist
Before upload label image is differentiated to high in the clouds, need that manually the image that scanner collects is rotated and cut, obtain
Standard picture, it is relatively complicated so to result in image processing process, and treatment effeciency is relatively low.
The content of the invention
The present invention provides a kind of method for correcting image and system, during solving label image processing in the prior art,
Need manually rotate the image that scanner collects and cut come image processing process caused by obtaining standard picture more
The problem of cumbersome, and treatment effeciency is relatively low.
Its specific technical solution is as follows：
A kind of method for correcting image, the described method includes：
When obtaining the original image that terminal uploads, place is zoomed in and out to the original image according to pre-set zoom condition
Reason, obtains input picture；
Determine each boundary line of the input picture, and the input picture vertex is determined according to the boundary line；
The apex coordinate of apex coordinate and the input picture in original image, determine original image with it is described
Correction parameter between input picture；
Judge whether the size of the input picture is more than pre-set dimension；
If so, positive correction then carries out the input picture according to the correction parameter；
If it is not, then the input picture is inversely corrected according to the correction parameter.
Optionally, each boundary line of the input picture is determined, including：
Determine the central point of the input picture, and the marginal point in four positive directions of the definite central point；
According to the marginal point and Hough algorithm, the pixel on same straight line is counted, and judges to unite
Whether the pixel sum of meter is more than predetermined threshold value；
If so, then the linear mark is boundary line；
If it is not, then abandon the straight line.
Optionally, the central point of the input picture, and the edge in four positive directions of the definite central point are determined
Point, including：
Traveled through with definite central point to positive direction, the RGB differences between the first pixel of acquisition and the second pixel
The sum of absolute value；
Judge whether the sum of absolute value of difference is more than predetermined threshold value；
If so, then using first pixel as marginal point.
Optionally, the input picture vertex is determined according to the border, including：
Determine J bars boundary line in each positive direction of four positive directions respectively in the boundary line of mark, wherein, J
For the positive integer more than or equal to 2；
According to the J bars boundary line determined, corresponding frame combination is generated；
Determine that coboundary and lower boundary ratio meet the frame combination of preset condition in frame combination；
Determine that the frame of closure degree maximum combines in the frame for meeting preset condition combines, and by the frame group
Conjunction is determined as label frame；
Using the intersection point in the boundary line of adjacent edge two-by-two of the label frame as corresponding apex coordinate.
Optionally, positive correction is carried out to the input picture, including：
First matrix and second matrix consistent with original image size are established, wherein, first matrix is used to tire out
Meter is mapped to the sum of pixel value each put in original image, and second matrix is used for the mapping amount for adding up corresponding points；
The sum of pixel value is being counted on into the first matrix, when pixel is counted on the second matrix, by first matrix
In the value each put divided by the second matrix in respective coordinates mapping amount, obtain the value of each pixel of positive correction.
A kind of image correction system, the system comprises：
Image pre-processing module, for obtain terminal upload original image when, according to pre-set zoom condition to described
Original image zooms in and out processing, obtains input picture；
Locating module, for determining each boundary line of the input picture, and determines according to the boundary line described defeated
Enter image vertex；
Correction module, the apex coordinate of apex coordinate and the input picture in original image, determines original
Correction parameter between image and the input picture；Judge whether the size of the input picture is more than pre-set dimension；If so,
Then according to the correction parameter, positive correction is carried out to the input picture, if it is not, then according to the correction parameter, to described
Input picture is inversely corrected.
Optionally, the locating module, the central point specifically for determining the input picture, and determine the central point
Four positive directions on marginal point；According to the marginal point and Hough algorithm, the pixel on same straight line is clicked through
Row statistics, and judge whether the pixel sum of statistics is more than predetermined threshold value；If so, then the linear mark is boundary line；If
It is no, then abandon the straight line.
Optionally, the locating module, specifically for being traveled through with definite central point to positive direction, obtains the first pixel
The sum of absolute value of RGB differences between the second pixel；Judge whether the sum of absolute value of difference is more than predetermined threshold value；If
It is, then using first pixel as marginal point.
Optionally, the locating module, specifically in the boundary line of mark respectively four positive directions it is each just
J bars boundary line is determined on direction, according to the J bars boundary line determined, generates corresponding frame combination；Combined in the frame
In determine coboundary and lower boundary ratio meets the frame of preset condition and combines；In the frame for meeting preset condition combines really
The frame combination of closure degree maximum is made, and frame combination is determined as label frame；By the two of the label frame
The intersection point in two adjacent edge boundary lines is as corresponding apex coordinate.
Optionally, the correction module, specifically for establishing first matrix and second consistent with original image size
Matrix, wherein, first matrix is used for the sum of accumulative pixel value each put being mapped in original image, second square
Battle array is used for the mapping amount for adding up corresponding points；The sum of pixel value is being counted on into the first matrix, pixel is counted on into the second square
During battle array, by the mapping amount of respective coordinates in the value each put in first matrix divided by the second matrix, positive school is obtained
The value of positive each pixel.
By the technical solution provided in the embodiment of the present invention, for terminal after original image is uploaded, system can be straight
Connect and master pattern is analyzed, then original image is positioned and corrected, it is accurate and quick, more intelligently, therefore scheme
The equipment of picture collection is no longer limited to scanner, can extend to more equipment such as mobile phone, digital camera.
In addition, the image after correction can be made to be standard picture by this method, eliminate manually rotation and cut
Operation, and processing accuracy improves a lot than manual operation, so systematic difference scope obtains larger extension so that
Improve the practicality of system.
Brief description of the drawings
Fig. 1 is a kind of flow chart of method for correcting image in the embodiment of the present invention；
Fig. 2 is the central point of input picture and the schematic diagram of four positive directions in the embodiment of the present invention；
Fig. 3 is that the pixel of positive correction in the embodiment of the present invention maps schematic diagram；
Fig. 4 is that the pixel inversely corrected in the embodiment of the present invention maps schematic diagram；
Fig. 5 is a kind of structure diagram of image correction system in the embodiment of the present invention.
Embodiment
Technical solution of the present invention is described in detail below by attached drawing and specific embodiment, it will be appreciated that this hair
Particular technique feature in bright embodiment and embodiment is the explanation to technical solution of the present invention, rather than is limited, not
In the case of conflict, the particular technique feature in the embodiment of the present invention and embodiment can be mutually combined.
It is a kind of flow chart of method for correcting image in the embodiment of the present invention as shown in Figure 1, this method includes：
S101, when obtaining the original image that terminal uploads, zooms in and out original image processing according to preset condition, obtains
To input picture；
In embodiments of the present invention, which can be mobile phone or other equipment with shooting function.Clapped in terminal
Take the photograph after image, terminal uploads onto the server the original image, and needing exist for explanation is, original image herein is straight for terminal
The image photographed is connect, and without the processing such as cutting by user.
After system obtains the original image, system will carry out size normalization to the original image, such as by original
The shorter edge of beginning image is uniformly reduced into 480 pixels, and longer sides are reduced, thus on year-on-year basis according to the diminution ratio of short side
It can obtain input picture.
S102, determines each boundary line of input picture, and determines input picture vertex according to boundary line；
After input picture is obtained, midpoint is determined in the input image, and four are then determined centered on the central point
A positive direction, i.e.,：The upper and lower, left and right of central point, than as shown in Figure 2.
After four positive directions are determined, system will carry out input picture endpoint detections, and specific detection mode is：With
Definite central point is traveled through to positive direction, obtains the sum of the absolute value of RGB differences between the first pixel and the second pixel,
Judge whether the sum of absolute value of difference is more than predetermined threshold value, if so, then using the first pixel as marginal point.
This is sentenced exemplified by therefrom heart point is upward to illustrate the endpoint detections method, the detection method phase on other directions
Together.Traversed up first from central point, if pixel P1(R1, G1, B1) differs+3 pixel with its y-coordinate, i.e.,：In P1
The point P of the position of poor 3 pixels in underface2The sum of absolute value of difference of RGB of (R2, G2, B2), more than threshold value T1, then by point
P1 is labeled as marginal point.Meet：
Sum(|Sub(R1,R2)|,|Sub(G1,G2)|,|Sub(B1,B2)|)>T1
The input picture marginal point can accurately be determined in the input image by this method.
After marginal point is determined, based on the marginal point detected, first using morphology operations, eliminate isolated in image
Point, and retain 8 neighborhoods adjacent pixel two-by-two.
Using Hough transformation, the pixel on same straight line is counted, and judge statistics pixel sum whether
More than predetermined threshold value, if so, being then boundary line by the linear mark, if it is not, then abandoning the straight line.
Further, in order to improve the accuracy of boundary line, so after the boundary line marked by Hough transformation, obtain
The angle of boundary line, i.e.,：Up-and-down boundary and horizontal direction angle, or right boundary and vertical direction angle, if the angle
More than predetermined threshold value, then the straight line is excluded.The straight line of partial invalidity can be filtered out by this way.
After marking all straight lines in the input image, respectively in each of four positive directions in the boundary line of mark
J bars boundary line is determined in positive direction, according to the J bars boundary line determined, corresponding frame combination is generated, in frame combination
Determine that coboundary and lower boundary ratio meet the frame combination of preset condition, determined in the frame for meeting preset condition combines
Go out the frame combination of closure degree maximum, and frame combination is determined as label frame.
For example, after straight line is obtained, from center is up and down, left and right four direction respectively retains 4 straight lines, four
Direction retains 16 straight lines altogether, this 16 straight lines will produce 256 kinds of frame combinations, in 256 kinds of frames combine, by coboundary
1 is kept off with the ratio of lower boundary:1 frame combination excludes, and the ratio of left margin and right margin is kept off 1:1 frame group
Close and exclude.Then the closure degree of each frame combination is determined in the combination of remaining frame, and determines closure degree maximum
Frame combination, by the frame combination be determined as label frame.
After definite label frame, according to the straight line parameter of the 4 of label frame boundary lines, adjacent boundary line two-by-two
Find intersection and can be obtained by corresponding apex coordinate.
Further, in embodiments of the present invention, since apex coordinate is in 480 diminution figure, 4, label is calculated
Apex coordinate, the coordinate of original image will be enlarged into according to scaling, while also error can be amplified, the amplification of its error
Multiple it is related with scaling, so after opposite vertexes coordinate is amplified, where vertex in the error range of frame region,
Edge detection is carried out again, edge is counted using least square method, is corrected the parameter of frame straight line, is recalculated vertex,
Error correction can be carried out with opposite vertexes coordinate by this way, so as to ensure that the accuracy of apex coordinate.
S103, the apex coordinate in apex coordinate and input picture in original image, determine original image with
Correction parameter between input picture；
In step s 102, each apex coordinate is determined in the input image, at this time can be according in original image
Apex coordinate in apex coordinate and input picture, establishes the coordinate mapping relations between input picture and original image.
For example, the upper and lower, left and right apex coordinate in input picture is respectively：
A(x1,y1)、B(x2,y2)、C(x3,y3)、D(x4,y4),
Upper and lower, left and right apex coordinate in original image is respectively：
A′(x1′,y1′)、B′(x′2,y′2)、C′(x3′,y3′)、D′(x′4,y′4),
Then according to following relational expression：
Wherein, x ', y ' are the apex coordinate in original image, and x, y are the apex coordinate in input picture.
So it is brought into above-mentioned apex coordinate is corresponding in above-mentioned relation formula, it is possible to solve k1、k2、k3、k4、k5、
k6、k7.Here k1、k2、k3、k4、k5、k6、k7As final correction parameter.
So after obtaining correction parameter, the arbitrary coordinate in input picture is brought into can be obtained by correction in above formula after
Coordinate.
S104, judges whether the size of input picture is more than or equal to pre-set dimension；
Correction parameter is being obtained, and before image rectification is carried out, it is also necessary to determine it is that forward direction is carried out to input picture
Correction or reverse correction.It can be determined in embodiments of the present invention according to picture size, that is to say, that when input picture
When size is more than pre-set dimension, then performs S105 and corrected using positive, if the size of input picture is less than pre-set dimension, then
S106 is performed using reverse correction.
Explanation is needed exist for, which can be the size of original image, naturally it is also possible to be other settings
Size.
S105, according to correction parameter, carries out input picture positive correction；
Specifically, first matrix and second matrix consistent with original image size are established, wherein, first square
Battle array is used for the sum of accumulative pixel value each put being mapped in original image, and second matrix is used to add up reflecting for corresponding points
Penetrate quantity；The sum of pixel value is being counted on into the first matrix, when pixel is counted on the second matrix, by first matrix
The value each put divided by the second matrix in respective coordinates mapping amount, obtain the value of each pixel of positive correction.
For example, corrected for forward direction, be to be mapped large-size images to small-sized image, as shown in figure 3, because
There to be multiple pixels to be mapped on a point of small-sized image in this large-sized image.Carrying out positive timing, meeting
Two matrix As consistent with original image size and B are established, matrix A is used to add up to be mapped to the picture that original image is each put
The sum of element value, matrix B are used for the mapping amount for adding up corresponding points.As shown in the figure, input picture has 4 points to be mapped to original image
On 1 point, the sum of pixel is stored in matrix A, and mapping points are stored in B, after the mapping for completing all the points, with each in matrix A
The value of point go divided by matrix B in respective coordinates mapping amount, that is, obtain the value of each pixel after positive correction.
S106, according to correction parameter, inversely corrects input picture.
For reverse correction mapped from large-size images to small-sized image, but picture is obtained from small-sized image
Element value, is filled up in large-size images, therefore may have multiple pixels to be mapped to the one of small-sized image in large-size images
On a point.Reverse timing is being carried out, using bilinear interpolation, is realizing many-to-one pixel filling.
As shown in figure 4, the point P (x, y) on original image passes through the transformation relation between picture size, input picture is mapped to
Point P ' on, the coordinate (x ', y ') of point P ' may be made of decimal.The coordinate of 4 pixel ABCD on input picture is known
And pixel value, the pixel value of P ' (x ', y ') is obtained according to bilinear interpolation formula, is filled into original image.
Bilinear interpolation is calculated by 3 parts and formed, and here first deposits the result of calculation of part 1 and part 2 temporarily
In tmp1And tmp2In, then substitute into third portion and continue to calculate.If pixel value is f.
The technical program is used in combination by positive correction and reverse two methods of correction, ensures the input for various situations
Image can quickly realize rational correction as much as possible, and effective guarantee is provided for next step graphical analysis.
By the technical solution provided in the embodiment of the present invention, for terminal after original image is uploaded, system can be straight
Connect and master pattern is analyzed, then original image is positioned and corrected, it is accurate and quick, more intelligently, therefore scheme
The equipment of picture collection is no longer limited to scanner, can extend to more equipment such as mobile phone, digital camera.
In addition, the image after correction can be made to be standard picture by this method, eliminate manually rotation and cut
Operation, and processing accuracy improves a lot than manual operation, so system application range obtains larger extension so that
Improve the practicality of system.
Further, the method that the embodiment of the present invention is provided is corresponded to, a kind of image calibration is additionally provided in the embodiment of the present invention
Positive system, a kind of structure diagram of image correction system, the system in the embodiment of the present invention of being illustrated in figure 5 include：
Image pre-processing module 501, for obtain terminal upload original image when, according to pre-set zoom condition to institute
State original image and zoom in and out processing, obtain input picture；
Locating module 502, for determining each boundary line of the input picture, and according to determining the boundary line
Input picture vertex；
Correction module 503, for the apex coordinate in original image and the apex coordinate of the input picture, really
Determine the correction parameter between original image and the input picture；Judge whether the size of the input picture is more than default ruler
It is very little；If so, positive correction then carries out the input picture according to the correction parameter, if it is not, then joining according to the correction
Number, inversely corrects the input picture.
Further, in embodiments of the present invention, the locating module 502, specifically for determining in the input picture
Heart point, and the marginal point in four positive directions of the definite central point；According to the marginal point and Hough algorithm, in
Pixel on same straight line is counted, and judges whether the pixel sum of statistics is more than predetermined threshold value；It is if so, then described
Linear mark is boundary line；If it is not, then abandon the straight line.
Further, in embodiments of the present invention, the locating module 502, specifically for definite central point to pros
To traversal, the sum of the absolute value of RGB differences between the first pixel of acquisition and the second pixel；Judge difference absolute value it
Whether predetermined threshold value is more than；If so, then using first pixel as marginal point.
Further, in embodiments of the present invention, the locating module 502, specifically for distinguishing in the boundary line of mark
J bars boundary line is determined in each positive direction of four positive directions, according to the J bars boundary line determined, generates corresponding side
Frame combines；Determine that coboundary and lower boundary ratio meet the frame combination of preset condition in frame combination；Meeting
The frame combination of closure degree maximum is determined in the frame combination of preset condition, and frame combination is determined as label edges
Frame；Using the intersection point in the boundary line of adjacent edge two-by-two of the label frame as corresponding apex coordinate.
Further, in embodiments of the present invention, the correction module 503, specifically for establishing and original image size one
The first matrix and the second matrix caused, wherein, first matrix is used to accumulative to be mapped to each putting in original image
The sum of pixel value, second matrix are used for the mapping amount for adding up corresponding points；The sum of pixel value is being counted on into the first matrix,
When pixel is counted on the second matrix, by respective coordinates in the value each put in first matrix divided by the second matrix
Mapping amount, obtains the value of each pixel of positive correction.
Although having been described for the preferred embodiment of the application, one of ordinary skilled in the art once knows substantially
Creative concept, then can make these embodiments other change and modification.So appended claims are intended to be construed to wrap
Include preferred embodiment and fall into all change and modification of the application scope, including vertex is determined using special symbol, mark
Deng change mode.
Obviously, those skilled in the art can carry out the application essence of the various modification and variations without departing from the application
God and scope.In this way, if these modifications and variations of the application belong to the scope of the application claim and its equivalent technologies
Within, then the application is also intended to comprising including these modification and variations.
Claims (10)
- A kind of 1. method for correcting image, it is characterised in that the described method includes：When obtaining the original image that terminal uploads, processing is zoomed in and out to the original image according to pre-set zoom condition, is obtained To input picture；Determine each boundary line of the input picture, and the input picture vertex is determined according to the boundary line；The apex coordinate of apex coordinate and the input picture in original image, determines original image and the input Correction parameter between image；Judge whether the size of the input picture is more than pre-set dimension；If so, positive correction then carries out the input picture according to the correction parameter；If it is not, then the input picture is inversely corrected according to the correction parameter.
- 2. the method as described in claim 1, it is characterised in that determine each boundary line of the input picture, including：Determine the central point of the input picture, and the marginal point in four positive directions of the definite central point；According to the marginal point and Hough algorithm, the pixel on same straight line is counted, and judges statistics Whether pixel sum is more than predetermined threshold value；If so, then the linear mark is boundary line；If it is not, then abandon the straight line.
- 3. the method as described in claim 1, it is characterised in that determine the central point of the input picture, and determine it is described in Marginal point in four positive directions of heart point, including：Traveled through with definite central point to positive direction, obtain the absolute of RGB differences between the first pixel and the second pixel The sum of value；Judge whether the sum of absolute value of difference is more than predetermined threshold value；If so, then using first pixel as marginal point.
- 4. method as claimed in claim 2, it is characterised in that the input picture vertex is determined according to the border, including：Determine J bars boundary line in each positive direction of four positive directions respectively in the boundary line of mark, wherein, J is big In the positive integer equal to 2；According to the J bars boundary line determined, corresponding frame combination is generated；Determine that coboundary and lower boundary ratio meet the frame combination of preset condition in frame combination；The frame combination of closure degree maximum is determined in the frame for meeting preset condition combines, and the frame is combined really It is set to label frame；Using the intersection point in the boundary line of adjacent edge two-by-two of the label frame as corresponding apex coordinate.
- 5. the method as described in claim 1, it is characterised in that positive correction is carried out to the input picture, including：First matrix and second matrix consistent with original image size are established, wherein, first matrix is used for accumulative reflect The sum of pixel value each put being mapped in original image, second matrix are used for the mapping amount for adding up corresponding points；The sum of pixel value is being counted on into the first matrix, when pixel is counted on the second matrix, by first matrix The mapping amount of respective coordinates in the value each put divided by the second matrix, obtains the value of each pixel of positive correction.
- A kind of 6. image correction system, it is characterised in that the system comprises：Image pre-processing module, for obtain terminal upload original image when, according to pre-set zoom condition to described original Image zooms in and out processing, obtains input picture；Locating module, the input figure is determined for determining each boundary line of the input picture, and according to the boundary line As vertex；Correction module, for the apex coordinate in original image and the apex coordinate of the input picture, determines original Correction parameter between image and the input picture；Judge whether the size of the input picture is more than pre-set dimension；If so, Then according to the correction parameter, positive correction is carried out to the input picture, if it is not, then according to the correction parameter, to described Input picture is inversely corrected.
- 7. system as claimed in claim 6, it is characterised in that the locating module, specifically for determining the input picture Central point, and determine the central point four positive directions on marginal point；It is right according to the marginal point and Hough algorithm Pixel on same straight line is counted, and judges whether the pixel sum of statistics is more than predetermined threshold value；If so, then The linear mark is boundary line；If it is not, then abandon the straight line.
- 8. system as claimed in claim 6, it is characterised in that the locating module, specifically for definite central point to Positive direction travels through, and obtains the sum of the absolute value of RGB differences between the first pixel and the second pixel；Judge the absolute of difference Whether the sum of value is more than predetermined threshold value；If so, then using first pixel as marginal point.
- 9. system as claimed in claim 6, it is characterised in that the locating module, specifically in the boundary line of mark J bars boundary line is determined in each positive direction of four positive directions respectively, according to the J bars boundary line determined, generation corresponds to Frame combination；Determine that coboundary and lower boundary ratio meet the frame combination of preset condition in frame combination； Meet the frame combination that closure degree maximum is determined in the frame combination of preset condition, and frame combination is determined as marking Sign frame；Using the intersection point in the boundary line of adjacent edge two-by-two of the label frame as corresponding apex coordinate.
- 10. system as claimed in claim 6, it is characterised in that the correction module, specifically for establishing and original image ruler Very little consistent the first matrix and the second matrix, wherein, first matrix is used to add up to be mapped to each in original image The sum of pixel value of point, second matrix are used for the mapping amount for adding up corresponding points；The sum of pixel value is being counted on first Matrix, will be corresponding in the value each put in first matrix divided by the second matrix when pixel is counted on the second matrix The mapping amount of coordinate, obtains the value of each pixel of positive correction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711043165.7A CN107909544B (en) | 2017-10-31 | 2017-10-31 | Image correction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711043165.7A CN107909544B (en) | 2017-10-31 | 2017-10-31 | Image correction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107909544A true CN107909544A (en) | 2018-04-13 |
| CN107909544B CN107909544B (en) | 2021-08-10 |
Family
ID=61842288
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711043165.7A Active CN107909544B (en) | 2017-10-31 | 2017-10-31 | Image correction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107909544B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108805124A (en) * | 2018-04-18 | 2018-11-13 | 北京嘀嘀无限科技发展有限公司 | Image processing method and device, computer readable storage medium |
| CN109949249A (en) * | 2019-03-29 | 2019-06-28 | 珠海丹德图像技术有限公司 | A kind of cylindrical picture bearing calibration and system |
| CN110356151A (en) * | 2019-05-28 | 2019-10-22 | 合肥晌玥科技有限公司 | A kind of automatic generation method and device in jade carving path |
| CN110390339A (en) * | 2019-07-16 | 2019-10-29 | 北京市计算中心 | A kind of method for correcting image, device and storage medium |
| CN110764764A (en) * | 2019-09-16 | 2020-02-07 | 平安科技（深圳）有限公司 | Webpage-side image fixing and stretching method and device, computer equipment and storage medium |
| CN115018868A (en) * | 2022-05-26 | 2022-09-06 | 贵州大学 | Computer image edge detection method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101197004A (en) * | 2007-12-25 | 2008-06-11 | 深圳矽感科技有限公司 | Two-dimension bar code and its coding and decoding method |
| CN101625760A (en) * | 2009-07-28 | 2010-01-13 | 谭洪舟 | Method for correcting certificate image inclination |
| JP4565396B2 (en) * | 2006-06-29 | 2010-10-20 | 富士ゼロックス株式会社 | Image processing apparatus and image processing program |
| CN201741188U (en) * | 2010-05-05 | 2011-02-09 | 深圳市优视技术有限公司 | Image self-adaptive scaling device capable of keeping edge smooth |
| US20140032406A1 (en) * | 2008-01-18 | 2014-01-30 | Mitek Systems | Systems for Mobile Image Capture and Remittance Processing of Documents on a Mobile Device |
| JP2014085693A (en) * | 2012-10-19 | 2014-05-12 | Alpine Electronics Inc | Lane monitoring system and lane monitoring method |
| CN107240064A (en) * | 2016-03-28 | 2017-10-10 | 展讯通信(天津)有限公司 | Method for correcting image, device and terminal |
-
2017
- 2017-10-31 CN CN201711043165.7A patent/CN107909544B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4565396B2 (en) * | 2006-06-29 | 2010-10-20 | 富士ゼロックス株式会社 | Image processing apparatus and image processing program |
| CN101197004A (en) * | 2007-12-25 | 2008-06-11 | 深圳矽感科技有限公司 | Two-dimension bar code and its coding and decoding method |
| US20140032406A1 (en) * | 2008-01-18 | 2014-01-30 | Mitek Systems | Systems for Mobile Image Capture and Remittance Processing of Documents on a Mobile Device |
| CN101625760A (en) * | 2009-07-28 | 2010-01-13 | 谭洪舟 | Method for correcting certificate image inclination |
| CN201741188U (en) * | 2010-05-05 | 2011-02-09 | 深圳市优视技术有限公司 | Image self-adaptive scaling device capable of keeping edge smooth |
| JP2014085693A (en) * | 2012-10-19 | 2014-05-12 | Alpine Electronics Inc | Lane monitoring system and lane monitoring method |
| CN107240064A (en) * | 2016-03-28 | 2017-10-10 | 展讯通信(天津)有限公司 | Method for correcting image, device and terminal |
Non-Patent Citations (1)
| Title |
|---|
| 钟金荣 等: "《特征提取和匹配的图像倾斜校正》", 《中国图象图形学报》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108805124A (en) * | 2018-04-18 | 2018-11-13 | 北京嘀嘀无限科技发展有限公司 | Image processing method and device, computer readable storage medium |
| CN108805124B (en) * | 2018-04-18 | 2019-10-18 | 北京嘀嘀无限科技发展有限公司 | Image processing method and device, computer readable storage medium |
| CN109949249A (en) * | 2019-03-29 | 2019-06-28 | 珠海丹德图像技术有限公司 | A kind of cylindrical picture bearing calibration and system |
| CN109949249B (en) * | 2019-03-29 | 2022-11-08 | 珠海丹德图像技术有限公司 | Cylindrical image correction method and system |
| CN110356151A (en) * | 2019-05-28 | 2019-10-22 | 合肥晌玥科技有限公司 | A kind of automatic generation method and device in jade carving path |
| CN110390339A (en) * | 2019-07-16 | 2019-10-29 | 北京市计算中心 | A kind of method for correcting image, device and storage medium |
| CN110764764A (en) * | 2019-09-16 | 2020-02-07 | 平安科技（深圳）有限公司 | Webpage-side image fixing and stretching method and device, computer equipment and storage medium |
| CN110764764B (en) * | 2019-09-16 | 2024-03-01 | 平安科技（深圳）有限公司 | Webpage end image fixed stretching method and device, computer equipment and storage medium |
| CN115018868A (en) * | 2022-05-26 | 2022-09-06 | 贵州大学 | Computer image edge detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107909544B (en) | 2021-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107909544A (en) | A kind of method for correcting image and system | |
| CN110390640B (en) | Template-based Poisson fusion image splicing method, system, equipment and medium | |
| CN110211043B (en) | Registration method based on grid optimization for panoramic image stitching | |
| CN105046657B (en) | A kind of image stretch distortion self-adapting correction method | |
| US7839422B2 (en) | Gradient-domain compositing | |
| CN103345736B (en) | A kind of virtual viewpoint rendering method | |
| CN106570436B (en) | A kind of substation equipment kilowatt meter reading-out system and its meter register method based on two dimensional code | |
| ES2438141T3 (en) | Device to restore images and method | |
| WO2017016050A1 (en) | Image preview method, apparatus and terminal | |
| CN103258316B (en) | Method and device for picture processing | |
| CN106981078B (en) | Sight line correction method and device, intelligent conference terminal and storage medium | |
| US20110228124A1 (en) | Character recognition preprocessing method and apparatus | |
| CN110400278A (en) | A kind of full-automatic bearing calibration, device and the equipment of color of image and geometric distortion | |
| CN114175091A (en) | Method for optimal body or face protection with adaptive dewarping based on context segmentation layer | |
| CN108269236B (en) | Image deviation rectifying method and device | |
| CN103295209A (en) | Splicing method and system for DR images | |
| CN108205822B (en) | Picture pasting method and device | |
| Islam et al. | Aesthetics-driven stereoscopic 3-D image recomposition with depth adaptation | |
| CN112085013A (en) | Tooth image acquisition device and image processing method | |
| CN104240259A (en) | High-speed photographing instrument voucher intelligent edge cutting and correcting system and high-speed photographing instrument voucher intelligent edge cutting and correcting method based on silhouette extraction | |
| CN109754381B (en) | Image processing method and system | |
| CN109949249B (en) | Cylindrical image correction method and system | |
| CN109472826A (en) | Localization method and device based on binocular vision | |
| CN101729739A (en) | Method for rectifying deviation of image | |
| CN106652898A (en) | LED display point-to-point correction method |
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 |