CN109698892A - A kind of video image sharpening method and image processing equipment - Google Patents
A kind of video image sharpening method and image processing equipment Download PDFInfo
- Publication number
- CN109698892A CN109698892A CN201811639745.7A CN201811639745A CN109698892A CN 109698892 A CN109698892 A CN 109698892A CN 201811639745 A CN201811639745 A CN 201811639745A CN 109698892 A CN109698892 A CN 109698892A
- Authority
- CN
- China
- Prior art keywords
- pixel
- value
- acutance
- level
- abs
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/14—Picture signal circuitry for video frequency region
- H04N5/20—Circuitry for controlling amplitude response
- H04N5/205—Circuitry for controlling amplitude response for correcting amplitude versus frequency characteristic
- H04N5/208—Circuitry for controlling amplitude response for correcting amplitude versus frequency characteristic for compensating for attenuation of high frequency components, e.g. crispening, aperture distortion correction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/60—Noise processing, e.g. detecting, correcting, reducing or removing noise
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
The present invention is suitable for field of image processing, provides video image sharpening method and image processing equipment.Method includes: that the pixel information of input is carried out capable caching, obtains the pixel of the n-quadrant N ﹡；By each pixel respectively multiplied by corresponding coefficient, and added up to obtain accumulated value；Central pixel point is also calculated separately in the acutance of multiple directions；Accumulated value is taken absolute value, different gains coefficient when different boundary is obtained, gain coefficient is obtained into new gain coefficient multiplied by gain adjustment value, by accumulated value multiplied by new gain coefficient, and the value after multiplication is added with archicenter pixel, obtains level-one sharpened value；Correlation Centre pixel is maximized in the acutance of multiple directions and sharpens gain as second level sharpened value multiplied by second level；Level-one sharpened value and second level sharpened value are added on the pixel of archicenter, obtain the sharpening result of central pixel point.Present invention sharpening effect is preferable, can also carry out stronger sharpening to weak boundary, for noise, can will sharpen gain and be pressed onto minimum, avoid the noise of excessive enlarged drawing.
Description
Technical field
The invention belongs to field of image processing more particularly to a kind of video image sharpening method and image processing equipments.
Background technique
The image information of video camera acquisition is passed due to the physical limit, working environment, image of imaging sensor itself
The factors such as the temperature influence of sensor, can include noise information.In the transmission and treatment process of video image, often to video
Image carries out the processing such as smothing filtering, although noise can be inhibited, the inevitably profile of fuzzy video image and thin
Section, causes the profile of video image and details unintelligible.The profile that can compensate for image by image sharpening enhances the side of image
Edge, details and Gray Level Jump part, become apparent from image.
The video image sharpening method of the prior art is mainly overlapped by fuzzy algorithmic approach obtain marginal information
It is preferable to retain, although these algorithms have preferable retention to the strong edge in image, to the weak side in image
Edge is unobvious.In addition, image sharpening, while enhancing the edge and details of video image, the noise of video image also can simultaneously
Excessively amplified, to influence the quality of video image.
Summary of the invention
The purpose of the present invention is to provide a kind of video image sharpening method, computer readable storage medium and image procossings
Equipment, it is intended to which the video image sharpening method for solving the prior art is unobvious to the weak edge in image, in enhancing video image
Edge and while details, the noise of video image also can excessively be amplified simultaneously, to influence the quality of video image
Problem.
In a first aspect, the present invention provides a kind of video image sharpening methods, which comprises
The pixel information of input is subjected to capable caching, obtains the pixel of the n-quadrant N ﹡, N is greater than 1 integer；
By each pixel in the n-quadrant N ﹡ respectively multiplied by corresponding coefficient, and will be multiplied by the area N ﹡ N after corresponding coefficient
All pixels point in domain is added up to obtain accumulated value, and accumulated value indicates the boundary information of central pixel point；Meanwhile also distinguishing
Calculate acutance of the central pixel point in multiple directions of the n-quadrant N ﹡；
Accumulated value is taken absolute value, and according to different gains coefficient when thoroughly deserving different boundary of accumulated value, it will
Gain coefficient obtains new gain coefficient multiplied by gain adjustment value, by accumulated value multiplied by new gain coefficient, and by after multiplication value with
The archicenter pixel in the region N*N is added, and obtains level-one sharpened value；Compare the central pixel point in the region N*N in multiple directions
Acutance, be maximized multiplied by second level sharpen gain as second level sharpened value；
Level-one sharpened value and second level sharpened value are added on the archicenter pixel of the n-quadrant N ﹡, are obtained in the n-quadrant N ﹡
The sharpening result of imago vegetarian refreshments.
Second aspect, the present invention provides a kind of computer readable storage medium, the computer readable storage medium is deposited
Computer program is contained, the step such as above-mentioned video image sharpening method is realized when the computer program is executed by processor
Suddenly.
The third aspect, the present invention provides a kind of image processing equipments, comprising:
One or more processors；
Memory；And
One or more computer programs, the processor and the memory are connected by bus, wherein one
Or multiple computer programs are stored in the memory, and are configured to be executed by one or more of processors,
The processor is realized when executing the computer program such as the step of above-mentioned video image sharpening method.
In the present invention, due to having carried out two-stage sharpening to pixel, it is preferable to sharpen effect；Again due to that will indicate center
The accumulated value of the boundary information of pixel takes absolute value, and according to different gains when thoroughly deserving different boundary of accumulated value
Coefficient can also carry out stronger sharpening to weak boundary to obtain level-one sharpened value, for noise, can will sharpen gain
It is pressed onto minimum, avoids the noise of excessive enlarged drawing.
Detailed description of the invention
Fig. 1 is the flow chart for the video image sharpening method that the embodiment of the present invention one provides.
Fig. 2 is fixed sharpening response curve schematic diagram.
Fig. 3 is adjustable sharpening response curve schematic diagram.
Fig. 4 is the specific block diagram for the image processing equipment that the embodiment of the present invention three provides.
Specific embodiment
In order to which the purpose of the present invention, technical solution and beneficial effect is more clearly understood, below in conjunction with attached drawing and implementation
Example, the present invention will be described in further detail.It should be appreciated that specific embodiment described herein is only used to explain this hair
It is bright, it is not intended to limit the present invention.
In order to illustrate technical solutions according to the invention, the following is a description of specific embodiments.
Embodiment one:
Referring to Fig. 1, the video image sharpening method that the embodiment of the present invention one provides is the following steps are included: should be noted
It is, if having substantially the same as a result, video image sharpening method of the invention is not limited with process sequence shown in FIG. 1.
S101, the pixel information of input is subjected to capable caching, obtains the pixel of the n-quadrant N ﹡, N is greater than 1 integer.
Pixel information can be pixel luminance signal or pixel carrier chrominance signal.
In the embodiment of the present invention one, N usually takes 3,5 or 7, naturally it is also possible to be other integers.Resultant effect, hardware disappear
Consumption etc., N is preferably 5.
Vacation lets N be 5, in the pixel in 5 region 5 ﹡, indicates the first row pixel with D11, D12, D13, D14, D15 respectively
Point, D21, D22, D23, D24, D25 indicate that the second row pixel, D31, D32, D33, D34, D35 indicate the third line pixel,
D41, D42, D43, D44, D45 indicate that fourth line pixel, D51, D52, D53, D54, D55 indicate fifth line pixel.
S102, by each pixel in the n-quadrant N ﹡ respectively multiplied by corresponding coefficient, and will be after multiplied by corresponding coefficient
All pixels point in the n-quadrant N ﹡ is added up to obtain accumulated value, and accumulated value indicates the boundary information of central pixel point；Meanwhile
Also calculate separately acutance of the central pixel point in multiple directions of the n-quadrant N ﹡.
In the embodiment of the present invention one, the central pixel point for calculating separately the n-quadrant N ﹡ has in the acutance of multiple directions
Body can be with are as follows:
The central pixel point of the n-quadrant N ﹡ is calculated separately in the acutance in 2 directions or the acutance in 4 directions, such as 4 directions
Acutance include the acutance in 0 degree of direction, the acutance in 45 degree of directions, the acutance in 90 degree of directions and 135 degree of directions acutance.Certainly
It can be other 4 directions.The acutance in 2 directions includes the acutance in 0 degree of direction and the acutance in 90 degree of directions, alternatively, 45 degree
The acutance of the acutance in direction and 135 degree of directions.
Calculate the n-quadrant N ﹡ central pixel point in one direction acutance specifically includes the following steps:
The central pixel point of the n-quadrant N ﹡ and two adjacent with the central pixel point pixel in the angular direction β are filtered
Wave, central pixel point, the first adjacent pixel and the corresponding filter factor of the second pixel are respectively f0, f1 and f2；
New central point pixel value=f0* center pixel point value+f1* the first pixel point value the second pixel of+f2* is calculated
Point value；
The acutance coefficient for being incorporated in each pixel of the angular direction β calculates new central point pixel value and in the angular direction β
The difference of other all pixels points takes the difference as acutance of the central pixel point in the angular direction β of the n-quadrant N ﹡, wherein in the angle side β
To the difference of acutance coefficient of each pixel be preferably 0.
Calculate the n-quadrant N ﹡ central pixel point in one direction acutance can also with specifically includes the following steps:
Be incorporated in each pixel of the angular direction β acutance coefficient calculate central point pixel value in the other of the angular direction β
The difference of all pixels point takes the difference as acutance of the central pixel point in the angular direction β of the n-quadrant N ﹡, wherein in the angular direction β
The difference of the acutance coefficient of each pixel is preferably 0.
In the embodiment of the present invention one, the corresponding coefficient of each pixel in the n-quadrant N ﹡ need to meet the following conditions: center
The coefficient of pixel and the pixel adjacent with central pixel point is positive, and the corresponding coefficient of central pixel point is maximum, distance
The remoter pixel of central pixel point, corresponding coefficient are smaller.Algorithms of different has different coefficients, for the pixel in 5 region 5 ﹡
Point, the corresponding coefficient of all pixels point in the region 5*5, which is added, is equal to 0.
Such as the pixel for 5 region 5 ﹡, D33 are central pixel points, the corresponding coefficient of the first row pixel is-
2, -4, -4, -4, -2, the corresponding coefficient of the second row pixel is -4,0,8,0, -4, and the third line pixel is corresponding
Coefficient is -4,8,24,8, -4, and the corresponding coefficient of fourth line pixel is -4,0,8,0, -4, and fifth line pixel is right respectively
The coefficient answered is -2, -4, -4, -4, -2.
In the embodiment of the present invention one, such as the central pixel point in 5 region 5 ﹡ is calculated in the acutance in 0 degree of direction, D31,
Five pixels of D32, D33, D34, D35, D32, D33, D34 are first filtered, and can retain D32's near D33 and D34
Information, the corresponding filter factor of D33, D32, D34 is f0, f1 and f2, such as f0, f1 and f2 are 16,8 and 8 respectively, are also possible to
Other values, so that new central point pixel value D33 ' _ 0=f0*D33+f1*D32+f2*D34 be calculated.Then it calculates newly
The difference of central point pixel value D33 ' and horizontal other pixels, the acutance coefficient of D33 ' are coef0, the acutance system of D31, D35
Number is coef1, and the acutance coefficient of D32, D34 are coef2, then 0 degree of directional sharpness can be expressed as df0=coef0*D33 '-
Coef1*D31-coef2*D32-coef2*D34-coef1*D35, wherein in the acutance coefficient of each pixel in 0 degree of direction
Difference preferably 0；Alternatively,
The central pixel point in 5 region 5 ﹡ is calculated in the acutance in 0 degree of direction, five pixels of D31, D32, D33, D34, D35
Then pixel centered on D33 is calculated the difference of central pixel point and horizontal other pixels, the acutance coefficient of D33 by point
For coef0, the acutance coefficient of D31, D35 are coef1, and the acutance coefficient of D32, D34 are coef2, then 0 degree of directional sharpness can be with table
It is shown as df0=coef0*D33-coef1*D31-coef2*D32-coef2*D34-coef1*D35, wherein in each of 0 degree of direction
The difference of the acutance coefficient of a pixel is preferably 0.
The central pixel point in 5 region 5 ﹡ is calculated in the acutance in 45 degree of directions, five pixels of D51, D42, D33, D24, D15
Point first filters D42, D33, D24, can retain the information of D42 and D24 near D33, the corresponding filtering system of D33, D42, D24
Number is f0, f1 and f2, such as f0, f1 and f2 are 16,8,8 respectively, are also possible to other values, and new central point pixel is calculated
Value D33 ' _ 45=f0*D33+f1*D42+f2*D24.Then it is other that new central point pixel value D33 ' _ 45 and 45 degree of directions are calculated
The difference of pixel, the acutance coefficient of D33 ' _ 45 are coef0, and the acutance coefficient of D51, D15 are coef1, the acutance of D42, D24
Coefficient is coef2, then 45 degree of directional sharpness can be expressed as df45=coef0*D33 ' _ 45-coef1*D51-coef1*D15-
Coef2*D42-coef2*D24, wherein the difference in the acutance coefficient of each pixel in 45 degree of directions is preferably 0；Alternatively,
The central pixel point in 5 region 5 ﹡ is calculated in the acutance in 45 degree of directions, five pixels of D51, D42, D33, D24, D15
Then pixel centered on D33 is calculated the difference of central pixel point and 45 degree of other pixels in direction, the acutance of D33 by point
Coefficient is coef0, and the acutance coefficient of D51, D15 are coef1, and the acutance coefficient of D42, D24 are coef2, then 45 degree of directional sharpness
It can be expressed as df45=coef0*D33-coef1*D51-coef1*D15-coef2*D42-coef2*D24, wherein at 45 degree
The difference of the acutance coefficient of each pixel in direction is preferably 0.
The central pixel point in 5 region 5 ﹡ is calculated in the acutance in 90 degree of directions, five pixels of D13, D23, D33, D43, D53
Point first filters D23, D33, D43, can retain the information of D23 and D43 near D33, the corresponding filtering system of D33, D23, D43
Number is f0, f1 and f2, such as f0, f1 and f2 are 2,1,1 respectively, are also possible to other values, and new central point pixel is calculated
Value D33 ' _ 90=f0*D33+f1*D23+f2*D43.Then new central point pixel value D33 ' _ 90 and vertical other pixels are calculated
The difference of point, the acutance coefficient of D33 ' _ 90 are coef0, the acutance coefficient coef1 of D13, D53, the acutance coefficient of D23, D43
Coef2, then 90 degree of directional sharpness can be expressed as df90=coef0*D33 ' _ 90-coef1*D13-coef1*D53-coef2*
D23-coef2*D43, wherein the difference in the acutance coefficient of each pixel in 90 degree of directions is preferably 0；Alternatively,
The central pixel point in 5 region 5 ﹡ is calculated in the acutance in 90 degree of directions, five pixels of D13, D23, D33, D43, D53
Then pixel centered on D33 is calculated the difference of central pixel point and 90 degree of other pixels in direction, the acutance of D33 by point
Coefficient is coef0, the acutance coefficient coef2 of the acutance coefficient coef1, D23, D43 of D13, D53, then 90 degree of directional sharpness can be with
It is expressed as df90=coef0*D33-coef1*D13-coef1*D53-coef2*D23-coef2*D43, wherein in 90 degree of directions
The difference of acutance coefficient of each pixel be preferably 0.
The central pixel point in 5 region 5 ﹡ is calculated in the acutance in 135 degree of directions, five pictures of D11, D22, D33, D44, D55
Vegetarian refreshments first filters D22, D33, D44, can retain the information of D22 and D44 near D33, the corresponding filtering of D33, D22, D44
Coefficient is f0, f1 and f2, such as f0, f1 and f2 are 1,2,1 respectively, are also possible to other values, and new central point picture is calculated
Plain value D33 ' _ 135=f0*D33+f1*D22+f2*D44.Then new central point pixel value D33 ' _ 135 and 135 degree of sides are calculated
To the difference of other pixels, the acutance coefficient of D33 ' _ 135 is coef0, the acutance coefficient coef1 of D11, D55, D22, D33's
Acutance coefficient coef2, then 135 degree of directional sharpness can be expressed as df135=coef0*D33 ' _ 135-coef1*D11-coef1*
D55-coef2*D22-coef2*D33, wherein the difference in the acutance coefficient of each pixel in 135 degree of directions is preferably 0, or
Person,
The central pixel point in 5 region 5 ﹡ is calculated in the acutance in 135 degree of directions, five pictures of D11, D22, D33, D44, D55
Then pixel centered on D33 is calculated the difference of central pixel point and 135 degree of other pixels in direction by vegetarian refreshments, D33's
Acutance coefficient is coef0, the acutance coefficient coef2 of the acutance coefficient coef1, D22, D33 of D11, D55, then 135 degree of directional sharpness
It can be expressed as df135=coef0*D33-coef1*D11-coef1*D55-coef2*D22-coef2*D33, wherein 135
The difference of the acutance coefficient for each pixel for spending direction is preferably 0.
S103, accumulated value is taken absolute value, and according to different gains system when thoroughly deserving different boundary of accumulated value
Number, obtains new gain coefficient multiplied by gain adjustment value for gain coefficient, by accumulated value multiplied by new gain coefficient, and will be after multiplication
Value is added with the archicenter pixel of the n-quadrant N ﹡, obtains level-one sharpened value；Compare the central pixel point of the n-quadrant N ﹡ multiple
The acutance in direction is maximized and sharpens gain as second level sharpened value multiplied by second level.
It is described to be specifically as follows according to different gains coefficient when thoroughly deserving different boundary of accumulated value:
Processing of tabling look-up is carried out to the absolute value of accumulated value, the table, which is that the absolute value of accumulated value is corresponding with gain coefficient, to close
It is table, the response curve of the table is as shown in Figure 2 and Figure 3.It is different that the response curve indicates that the boundary to varying strength, profile carry out
The Edge contrast of intensity.Mapping table can also be adjusted according to practical application, adjust different sharpness, can also to weak boundary
To carry out stronger sharpening, and to very noisy, gain will be sharpened and be pressed onto minimum.From the figure it may be seen that on non-boundary or boundary
When information itself is very strong, sharpening coefficient very little, when boundary information is weaker or not strong, sharpening coefficient is larger, can promote this portion
Divide boundary information.Furthermore, it is possible to adjust response curve, part boundary information is sharpened to adjust.In addition, due to noise
Brightness is generally present in compared with the dark noise of dark scene or boundary obviously white noise, to both noises, sharpening coefficient
Very little will not exceedingly amplify noise.
In the embodiment of the present invention one, when new gain coefficient is greater than preset peak, the value of new gain coefficient is adjusted
Less than preset peak.
The central pixel point of the relatively n-quadrant N ﹡ is maximized in the acutance of multiple directions and sharpens increasing multiplied by second level
Benefit is specifically as follows as second level sharpened value: comparing the central pixel point of the n-quadrant N ﹡ in the acutance in 4 directions, is maximized again
Gain is sharpened as second level sharpened value multiplied by second level.
The central pixel point of the relatively n-quadrant N ﹡ is maximized in the acutance in 4 directions and sharpens gain multiplied by second level
It is specifically as follows as second level sharpened value:
Obtaining 0 degree of directional sharpness df0,45 degree of directional sharpness df45,90 degree of directional sharpness df90 and 135 degree of directional sharpness
It after df135, takes absolute value be denoted as abs_df0, abs_df45, abs_df90 and abs_df135 respectively, seek abs_df0-abs_
The value of df90, abs_df45-abs_df135 simultaneously take its absolute value, are denoted as abs_dlt1, abs_dlt2 respectively, compare abs_
The size of dlt1, abs_dlt2 continue the size for comparing abs_df0, abs_df90 if abs_dlt1 is larger, if
Abs_df0 is larger, then takes df0 to sharpen gain as second level sharpened value multiplied by second level, if abs_df90 is larger, take df90
Gain is sharpened as second level sharpened value multiplied by second level；If abs_dlt2 is larger, continue to compare abs_df45, abs_
The size of df135 takes df90 to sharpen gain as second level sharpened value multiplied by second level if abs_df90 is larger, if
Abs_df135 is larger, then df135 is taken to sharpen gain as second level sharpened value multiplied by second level.
S104, level-one sharpened value and second level sharpened value are added on the archicenter pixel of the n-quadrant N ﹡, obtain the area N ﹡ N
The sharpening result of the central pixel point in domain.
Embodiment two:
Second embodiment of the present invention provides a kind of computer readable storage medium, the computer-readable recording medium storage
There is computer program, realizes that the video image provided such as the embodiment of the present invention one is sharp when the computer program is executed by processor
The step of change method.
Embodiment three:
Fig. 4 shows the specific block diagram of the image processing equipment of the offer of the embodiment of the present invention three, a kind of image procossing
Equipment 100 includes: one or more processors 101, memory 102 and one or more computer programs, wherein the place
Reason device 101 is connected with the memory 102 by bus, and one or more of computer programs are stored in the memory
It in 102, and is configured to be executed by one or more of processors 101, the processor 101 executes the computer journey
The step of video image sharpening method such as the offer of the embodiment of the present invention one is realized when sequence.
The image processing equipment that the embodiment of the present invention three provides can be video camera etc..
In the present invention, due to having carried out two-stage sharpening to pixel, it is preferable to sharpen effect；Again due to that will indicate center
The accumulated value of the boundary information of pixel takes absolute value, and according to different gains when thoroughly deserving different boundary of accumulated value
Coefficient can also carry out stronger sharpening to weak boundary to obtain level-one sharpened value, for noise, can will sharpen gain
It is pressed onto minimum, avoids the noise of excessive enlarged drawing.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random
Access Memory), disk or CD etc..
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of video image sharpening method, which is characterized in that the described method includes:
The pixel information of input is subjected to capable caching, obtains the pixel of the n-quadrant N ﹡, N is greater than 1 integer；
By each pixel in the n-quadrant N ﹡ respectively multiplied by corresponding coefficient, and will be multiplied by the n-quadrant N ﹡ after corresponding coefficient
All pixels point added up to obtain accumulated value, accumulated value indicates the boundary information of central pixel point；Meanwhile it also calculating separately
Acutance of the central pixel point of the n-quadrant N ﹡ in multiple directions；
Accumulated value is taken absolute value, and according to different gains coefficient when thoroughly deserving different boundary of accumulated value, by gain
Coefficient obtains new gain coefficient multiplied by gain adjustment value, by accumulated value multiplied by new gain coefficient, and by the area Zhi Yu N*N after multiplication
The archicenter pixel in domain is added, and obtains level-one sharpened value；Compare the central pixel point in the region N*N in the sharp of multiple directions
Degree is maximized and sharpens gain as second level sharpened value multiplied by second level；
Level-one sharpened value and second level sharpened value are added on the archicenter pixel of the n-quadrant N ﹡, obtain the middle imago of the n-quadrant N ﹡
The sharpening result of vegetarian refreshments.
2. the method as described in claim 1, which is characterized in that the pixel information is pixel luminance signal or pixel
Point carrier chrominance signal；N takes 3,5 or 7.
3. the method as described in claim 1, which is characterized in that the central pixel point for calculating separately the n-quadrant N ﹡ is multiple
The acutance in direction specifically:
The central pixel point of the n-quadrant N ﹡ is calculated separately in the acutance in 2 directions or the acutance in 4 directions；
The central pixel point of the relatively n-quadrant N ﹡ is maximized in the acutance of multiple directions and sharpens gain work multiplied by second level
For second level sharpened value specifically: compare the central pixel point of the n-quadrant N ﹡ in the acutance in 2 directions or the acutance in 4 directions, take most
Big value sharpens gain as second level sharpened value multiplied by second level.
4. method as claimed in claim 3, which is characterized in that the acutance in 4 directions includes the acutance in 0 degree of direction, 45
Spend the acutance of the acutance in direction, the acutance in 90 degree of directions and 135 degree of directions；The acutance in 2 directions includes the sharp of 0 degree of direction
The acutance of degree and 90 degree of directions, alternatively, the acutance of the acutance in 45 degree of directions and 135 degree of directions.
5. the method as described in claim 1, which is characterized in that the corresponding coefficient of each pixel in the n-quadrant the N ﹡ needs
Meet the following conditions: the coefficient of central pixel point and the pixel adjacent with central pixel point is positive, and central pixel point pair
The coefficient answered is maximum, the remoter pixel of distance center pixel, and corresponding coefficient is smaller.
6. the method as described in claim 1, which is characterized in that described according to accumulated value when thoroughly deserving different boundary
Different gains coefficient specifically:
Processing of tabling look-up is carried out to the absolute value of accumulated value, the table is the absolute value of accumulated value and the corresponding relationship of gain coefficient
Table.
7. the method as described in claim 1, which is characterized in that the central pixel point for calculating the n-quadrant N ﹡ is sharp in one direction
Degree specifically includes:
The central pixel point of the n-quadrant N ﹡ and two adjacent with the central pixel point pixel in the angular direction β are filtered, in
Imago vegetarian refreshments, the first adjacent pixel and the corresponding filter factor of the second pixel are respectively f0, f1 and f2；
New central point pixel value=f0* center pixel point value+f1* the first pixel point value the second pixel of+f2* is calculated
Value；
Be incorporated in each pixel of the angular direction β acutance coefficient calculate new central point pixel value in the other of the angular direction β
The difference of all pixels point takes the difference as acutance of the central pixel point in the angular direction β of the n-quadrant N ﹡, wherein in the angular direction β
The acutance coefficient of each pixel and be 0；
Alternatively,
Acutance specifically includes the central pixel point of the calculating n-quadrant N ﹡ in one direction:
The acutance coefficient for being incorporated in each pixel of the angular direction β calculates central point pixel value to be owned in the other of the angular direction β
The difference of pixel takes the difference as acutance of the central pixel point in the angular direction β of the n-quadrant N ﹡.
8. method as claimed in claim 4, which is characterized in that the central pixel point of the relatively n-quadrant N ﹡ is in 4 directions
Acutance is maximized and sharpens gain as second level sharpened value multiplied by second level specifically:
Obtaining 0 degree of directional sharpness df0,45 degree of directional sharpness df45,90 degree of directional sharpness df90 and 135 degree of directional sharpness
It after df135, takes absolute value be denoted as abs_df0, abs_df45, abs_df90 and abs_df135 respectively, seek abs_df0-abs_
The value of df90, abs_df45-abs_df135 simultaneously take its absolute value, are denoted as abs_dlt1, abs_dlt2 respectively, compare abs_
The size of dlt1, abs_dlt2 continue the size for comparing abs_df0, abs_df90 if abs_dlt1 is larger, if
Abs_df0 is larger, then takes df0 to sharpen gain as second level sharpened value multiplied by second level, if abs_df90 is larger, take df90
Gain is sharpened as second level sharpened value multiplied by second level；If abs_dlt2 is larger, continue to compare abs_df45, abs_
The size of df135 takes df90 to sharpen gain as second level sharpened value multiplied by second level if abs_df90 is larger, if
Abs_df135 is larger, then df135 is taken to sharpen gain as second level sharpened value multiplied by second level.
9. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In the computer program realizes video image sharpening method as claimed in any one of claims 1 to 8 when being executed by processor
The step of.
10. a kind of image processing equipment, comprising:
One or more processors；
Memory；And
One or more computer programs, the processor and the memory are connected by bus, wherein one or more
A computer program is stored in the memory, and is configured to be executed by one or more of processors, special
Sign is that the processor realizes that video image as claimed in any one of claims 1 to 8 is sharp when executing the computer program
The step of change method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811639745.7A CN109698892B (en) | 2018-12-29 | 2018-12-29 | Video image sharpening method and image processing equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811639745.7A CN109698892B (en) | 2018-12-29 | 2018-12-29 | Video image sharpening method and image processing equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109698892A true CN109698892A (en) | 2019-04-30 |
| CN109698892B CN109698892B (en) | 2021-06-11 |
Family
ID=66232656
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811639745.7A Active CN109698892B (en) | 2018-12-29 | 2018-12-29 | Video image sharpening method and image processing equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109698892B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110942440A (en) * | 2019-12-12 | 2020-03-31 | 上海富瀚微电子股份有限公司 | Image sharpening method and device |
| CN113610727A (en) * | 2021-08-05 | 2021-11-05 | 上海交通大学 | Boundary reconstruction sharpening method for two-phase volume fraction image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050111055A1 (en) * | 2003-11-21 | 2005-05-26 | Seung-Joon Yang | Apparatus and method of image sharpness enhancement |
| CN102137221A (en) * | 2010-01-22 | 2011-07-27 | 富士通微电子(上海)有限公司 | Method and system for sharpening digital video image |
| CN105046655A (en) * | 2015-06-24 | 2015-11-11 | 武汉烽火众智数字技术有限责任公司 | Method and device for automatically sharpening video image |
-
2018
- 2018-12-29 CN CN201811639745.7A patent/CN109698892B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050111055A1 (en) * | 2003-11-21 | 2005-05-26 | Seung-Joon Yang | Apparatus and method of image sharpness enhancement |
| CN102137221A (en) * | 2010-01-22 | 2011-07-27 | 富士通微电子(上海)有限公司 | Method and system for sharpening digital video image |
| CN105046655A (en) * | 2015-06-24 | 2015-11-11 | 武汉烽火众智数字技术有限责任公司 | Method and device for automatically sharpening video image |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110942440A (en) * | 2019-12-12 | 2020-03-31 | 上海富瀚微电子股份有限公司 | Image sharpening method and device |
| CN110942440B (en) * | 2019-12-12 | 2023-04-18 | 上海富瀚微电子股份有限公司 | Image sharpening method and device |
| CN113610727A (en) * | 2021-08-05 | 2021-11-05 | 上海交通大学 | Boundary reconstruction sharpening method for two-phase volume fraction image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109698892B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2016206087A1 (en) | Low-illumination image processing method and device | |
| US8131104B2 (en) | Method and apparatus for adjusting the contrast of an input image | |
| WO2010084431A1 (en) | Method and apparatus for generating enhanced images | |
| JP6587317B2 (en) | Guided filter-based detail enhancement | |
| CN109584198B (en) | Method and device for evaluating quality of face image and computer readable storage medium | |
| CN110503617B (en) | Underwater image enhancement method based on high-frequency and low-frequency information fusion | |
| CN112419162A (en) | Image defogging method and device, electronic equipment and readable storage medium | |
| US9401009B2 (en) | Method and apparatus for enhancing quality of 3D image | |
| EP2059902A1 (en) | Method and apparatus for image enhancement | |
| KR102195047B1 (en) | Method and apparatus for enhancing quality of 3D image | |
| CN109698892A (en) | A kind of video image sharpening method and image processing equipment | |
| CN110942427A (en) | Image noise reduction method and device, equipment and storage medium | |
| KR102106073B1 (en) | Method and apparatus for processing images | |
| US11222412B2 (en) | Image processing device and method | |
| Chang | Entropy-based trilateral filtering for noise removal in digital images | |
| US20100002952A1 (en) | Method and apparatus for image sharpening | |
| CN106027853B (en) | Image processing apparatus, photographic device and image processing method | |
| CN112334942A (en) | Image processing method and device | |
| Chen et al. | Texture-preserving image deblurring | |
| CN114897735A (en) | Local self-adaptive underwater image contrast enhancement method | |
| CN111242876B (en) | Low contrast image enhancement method, apparatus and computer readable storage medium | |
| CN115689906A (en) | Image processing method and device | |
| Sharma et al. | Synthesis of flash and no-flash image pairs using guided image filtering | |
| Chang et al. | Rician noise removal in MR images using an adaptive trilateral filter | |
| Hast et al. | Retrospective illumination correction of greyscale historical aerial photos |
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 |