CN107301627A - Dynamic image unsharp masking removes artifact Enhancement Method and device - Google Patents
Dynamic image unsharp masking removes artifact Enhancement Method and device Download PDFInfo
- Publication number
- CN107301627A CN107301627A CN201710492125.4A CN201710492125A CN107301627A CN 107301627 A CN107301627 A CN 107301627A CN 201710492125 A CN201710492125 A CN 201710492125A CN 107301627 A CN107301627 A CN 107301627A
- Authority
- CN
- China
- Prior art keywords
- artifact
- mask
- negative edge
- mrow
- edge region
- 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
- 230000000873 masking effect Effects 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 25
- 230000006835 compression Effects 0.000 claims abstract description 94
- 238000007906 compression Methods 0.000 claims abstract description 94
- 230000002708 enhancing effect Effects 0.000 claims abstract description 85
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 15
- 238000003786 synthesis reaction Methods 0.000 claims abstract description 15
- 238000006243 chemical reaction Methods 0.000 claims description 24
- 230000003628 erosive effect Effects 0.000 claims description 16
- 238000000605 extraction Methods 0.000 claims description 14
- 238000005260 corrosion Methods 0.000 claims description 8
- 230000007797 corrosion Effects 0.000 claims description 8
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 11
- 238000010586 diagram Methods 0.000 description 7
- 238000000354 decomposition reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 239000000686 essence Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
- G06T5/75—Unsharp masking
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The present invention goes artifact Enhancement Method suitable for field of medical image processing there is provided a kind of dynamic image unsharp masking, including：Pending dynamic image is decomposed using unsharp masking USM enhancing algorithms, high frequency detail image and low frequency contour images is obtained；Generating details enhancing according to high frequency detail image goes artifact mask and dynamic range compression to remove artifact mask；Artifact mask and dynamic range compression is gone to go artifact mask to carry out going the enhancing of artifact details to handle to high frequency detail image using details enhancing, obtain the enhanced high frequency detail image of artifact details, dynamic range compression, the low frequency contour images after being compressed are carried out to low frequency contour images；The low frequency contour images that synthesis is gone after the enhanced high frequency detail image of artifact details and compression, obtain the dynamic image of pseudo- movie queen；The method that the present invention is provided is so that high frequency detail image carries out going artifact to optimize while details enhancing is carried out, to eliminate the artifact of generation, raising dynamic image display effect.
Description
Technical field
Artifact enhancing side is gone to the invention belongs to field of medical image processing, more particularly to a kind of dynamic image unsharp masking
Method and device.
Background technology
USM (Unsharp Mask, unsharp masking) strengthens algorithm, is usually used in strengthening dynamic image, to improve
Dynamic image display effect.USM enhancing the main of algorithm realizes that thought is：Fuzzy Processing is carried out to original dynamic image and obtains low
Frequency contour images, and high frequency detail image is obtained as difference with original dynamic image and the low frequency figure, so as to isolate high frequency detail
Image and low frequency contour images, then carry out details enhancing to high frequency detail image, and dynamic range is carried out to low frequency contour images
Compression, and the low frequency contour images after compression and enhanced high frequency detail image addition are obtained final enhanced image.
But strengthen algorithm using USM and carry out the composograph after details enhancing and dynamic range compression near strong edge
Artifact occurs, main cause there are two aspects：(1) nearby high frequency detail image enhaucament is excessive for strong edge；(2) near strong edge
The dynamic range compression of low frequency contour images, makes the relative enhancing of high frequency detail image excessive.Artifact can have a strong impact on the sense of image
See, diagnosis of the doctor to focus can be influenceed when medical science dynamic image strengthens.
The content of the invention
The present invention provides a kind of dynamic image unsharp masking and removes artifact Enhancement Method and device, it is desirable to provide one kind goes puppet
The method of shadow strengthens the artifact produced when algorithm is handled dynamic image to eliminate USM, improves dynamic image display effect.
Artifact Enhancement Method is gone the invention provides a kind of dynamic image unsharp masking, including：
Pending dynamic image is decomposed using unsharp masking USM enhancing algorithms, high frequency detail image is obtained
With low frequency contour images；
Generating details enhancing according to the high frequency detail image goes artifact mask and dynamic range compression to remove artifact mask；
Artifact mask and dynamic range compression is gone to remove artifact mask to the high frequency detail image using details enhancing
Progress goes the enhancing of artifact details to handle, and the enhanced high frequency detail image of artifact details is obtained, to the low frequency contour images
Carry out dynamic range compression, the low frequency contour images after being compressed；
The low frequency contour images that synthesis is gone after the enhanced high frequency detail image of artifact details and compression, obtain pseudo- movie queen
Dynamic image.
Further, generating details enhancing according to the high frequency detail image goes artifact mask and dynamic range compression to go puppet
Shadow mask, including：
Binary conversion treatment is carried out to the high frequency detail image, negative edge region mask is obtained；
The information in negative edge region is extracted according to the negative edge region mask, and the information in the negative edge region is entered
Row normalized, obtains the information in the negative edge region after normalized；
Erosion operation is done to the information in the negative edge region after normalized with reference to contiguous range set in advance, and expanded
Big negative edge regional extent, obtains expanding the information in the negative edge region of negative edge regional extent；
The information in the negative edge region to expanding negative edge regional extent enters line translation stretching, obtains after conversion stretching
Negative edge region information；
Artifact mask is removed using the information generation in the negative edge region after obtained conversion stretching；
The acquisition details enhancing of artifact mask is gone to go artifact mask and dynamic range compression to remove artifact mask according to described.
Further, the negative edge region mask is：
The formula for extracting the information in negative edge region is：
NegHigh=NegMask*HighImg,
The formula of normalized is：
The formula of erosion operation is：
ErodeNegHigh=Erode (NegHigh, NeibSize),
Expand negative edge regional extent formula be：
NegHigh=NegMask*ErodeNegHigh+ (1-NegMask) * NegHigh；
Converting the formula stretched is：
NegHigh=NegHighgamma,
Generation goes the formula of artifact mask to be：
DeArtiMask=1-deArtiParam+deArtiParam*NegHigh,
Obtaining details enhancing goes the formula of artifact mask to be：
EeDeArtiMask=1+eeParam*DeArtiMask,
Obtaining dynamic range compression goes the formula of artifact mask to be：
DrcDeArtiMask=drcParam+ (1-drcParam) * DeArtiMask,
Wherein, NegMask is negative edge region mask, and HighImg (r, c) is that coordinate is (r, c) in high frequency detail image
Point pixel value, r represents row coordinate, and c represents row coordinate；NegHigh is the information in negative edge region；MinVal is represented
NegHigh minimum value, MaxVal represents NegHigh maximum；ErodeNegHigh is the negative edge region after corrosion
Information, Erode () is that the information in negative edge region is corroded, and NeibSize represents contiguous range set in advance；
Gamma spans are 3~7；DeArtiMask is removes artifact mask, and deArtiParam is artifact strength, and span is
0.0~1.0；EeParam is that details strengthens intensive parameter, and EeDeArtiMask is that artifact mask is removed in details enhancing；drcParam
For dynamic range compression parameter, DrcDeArtiMask is that dynamic range compression removes artifact mask.
Further, it is described to go artifact mask and dynamic range compression to remove artifact mask to described using details enhancing
High frequency detail image carries out going the enhancing of artifact details to handle, and obtains the enhanced high frequency detail image of artifact details, including：
Being multiplied by the details enhancing using high frequency detail image goes artifact mask and the dynamic range compression to go artifact to cover
Mould, obtains the enhanced high frequency detail image of artifact details；
It is described to the low frequency contour images carry out dynamic range compression, the low frequency contour images after being compressed, including：
Preset dynamic range compression parameter is multiplied by using low frequency contour images, the low frequency contour images after being compressed.
Further, the low frequency profile diagram that the synthesis is gone after the enhanced high frequency detail image of artifact details and compression
Picture, obtains the dynamic image of pseudo- movie queen, including：
Go the enhanced high frequency detail image of artifact details to be added with the low frequency contour images after the compression by described, obtain
To the dynamic image for going pseudo- movie queen.
Artifact intensifier is removed present invention also offers a kind of dynamic image unsharp masking, including：
Decomposing module, for being decomposed using unsharp masking USM enhancing algorithms to pending dynamic image, is obtained
High frequency detail image and low frequency contour images；
Artifact mask generating module is removed, artifact mask is removed and dynamic for generating details enhancing according to the high frequency detail image
State Ratage Coutpressioit removes artifact mask；
Processing module, for going artifact mask and dynamic range compression to remove artifact mask to described using details enhancing
High frequency detail image carries out going the enhancing of artifact details to handle, and the enhanced high frequency detail image of artifact details is obtained, to described
Low frequency contour images carry out dynamic range compression, the low frequency contour images after being compressed；
Synthesis module, for synthesizing the low frequency profile diagram gone after the enhanced high frequency detail image of artifact details and compression
Picture, obtains the dynamic image of pseudo- movie queen.
Further, it is described to go artifact mask generating module to specifically include：Negative edge region mask extraction module, negative edge
Area information extraction module, negative edge region extension module, conversion stretching module, the first mask generating module and the life of the second mask
Into module；
The negative edge region mask extraction module, for carrying out binary conversion treatment to the high frequency detail image, is obtained
Negative edge region mask；
The negative edge area information extraction module, for extracting negative edge region according to the negative edge region mask
Information, and the information in the negative edge region is normalized, obtain the letter in the negative edge region after normalized
Breath；
Negative edge region extension module, for combining contiguous range set in advance to the negative side after normalized
The information in edge region does erosion operation, and expands negative edge regional extent, obtains expanding the negative edge area of negative edge regional extent
The information in domain；
The conversion stretching module, the information for the negative edge region to expanding negative edge regional extent becomes
Stretching is changed, the information in the negative edge region after conversion stretching is obtained；
First mask generating module, goes for the information generation using the negative edge region after obtained conversion stretching
Artifact mask；
Second mask generating module, artifact mask is removed and dynamic for going artifact mask to obtain details enhancing according to
State Ratage Coutpressioit removes artifact mask.
Further, the negative edge region mask is：
The formula for extracting the information in negative edge region is：
NegHigh=NegMask*HighImg,
The formula of normalized is：
The formula of erosion operation is：
ErodeNegHigh=Erode (NegHigh, NeibSize),
Expand negative edge regional extent formula be：
NegHigh=NegMask*ErodeNegHigh+ (1-NegMask) * NegHigh；
Converting the formula stretched is：
NegHigh=NegHighgamma,
Generation goes the formula of artifact mask to be：
DeArtiMask=1-deArtiParam+deArtiParam*NegHigh,
Obtaining details enhancing goes the formula of artifact mask to be：
EeDeArtiMask=1+eeParam*DeArtiMask,
Obtaining dynamic range compression goes the formula of artifact mask to be：
DrcDeArtiMask=drcParam+ (1-drcParam) * DeArtiMask,
Wherein, NegMask is negative edge region mask, and HighImg (r, c) is that coordinate is (r, c) in high frequency detail image
Point pixel value, r represents row coordinate, and c represents row coordinate；NegHigh is the information in negative edge region；MinVal is represented
NegHigh minimum value, MaxVal represents NegHigh maximum；ErodeNegHigh is the negative edge region after corrosion
Information, Erode () is that the information in negative edge region is corroded, and NeibSize represents contiguous range set in advance；
Gamma spans are 3~7；DeArtiMask is removes artifact mask, and deArtiParam is artifact strength, and span is
0.0~1.0；EeParam is that details strengthens intensive parameter, and EeDeArtiMask is that artifact mask is removed in details enhancing；drcParam
For dynamic range compression parameter, DrcDeArtiMask is that dynamic range compression removes artifact mask.
Further, the processing module includes：First processing module and Second processing module；
The first processing module, artifact mask is removed and described for being multiplied by the details enhancing using high frequency detail image
Dynamic range compression removes artifact mask, obtains the enhanced high frequency detail image of artifact details；
The Second processing module, for being multiplied by preset dynamic range compression parameter using low frequency contour images, is obtained
Low frequency contour images after compression.
Further, the synthesis module, specifically for by it is described go the enhanced high frequency detail image of artifact details and
Low frequency contour images after the compression are added, and obtain the dynamic image of pseudo- movie queen.
Compared with prior art, beneficial effect is the present invention：A kind of dynamic image unsharp masking that the present invention is provided
Artifact Enhancement Method and device are removed, pending dynamic image is decomposed using unsharp masking USM enhancing algorithms, obtained
High frequency detail image and low frequency contour images；Details enhancing is generated according to the high frequency detail image and removes artifact mask and dynamic model
Artifact mask is removed in confined pressure contracting；And go artifact mask and dynamic range compression to remove artifact mask to high frequency using the details enhancing of generation
Detail pictures carry out going the enhancing of artifact details to handle, and the enhanced high frequency detail image of artifact details are obtained, to the low frequency
Contour images carry out dynamic range compression, the low frequency contour images after being compressed；And the enhanced height of artifact details is removed in synthesis
Low frequency contour images after frequency detail pictures and compression, obtain the dynamic image of pseudo- movie queen；The present invention compared with prior art,
It is that generation details enhancing goes artifact mask and dynamic range compression to remove artifact mask the characteristics of causing artifact according to strong edge enhancing,
Using USM enhancing algorithm pending dynamic image is handled when, using generation details enhancing go artifact mask with
Dynamic range compression goes artifact mask to handle high frequency detail image so that high frequency detail image is enhanced in progress details
Carry out going artifact to optimize simultaneously, to eliminate the artifact of generation, improve dynamic image display effect.
Brief description of the drawings
Fig. 1 is the implementation process that a kind of dynamic image unsharp masking provided in an embodiment of the present invention goes artifact Enhancement Method
Schematic diagram；
Fig. 2 is that a kind of dynamic image unsharp masking provided in an embodiment of the present invention goes the flow of artifact Enhancement Method to show
It is intended to；
Fig. 3 a are provided in an embodiment of the present invention to utilize original USM enhancing algorithms to carry out details enhancing and dynamic range compression
Image effect schematic diagram after processing；
Fig. 3 b be it is provided in an embodiment of the present invention using technical scheme provided in an embodiment of the present invention carry out go artifact to optimize
Effect diagram afterwards；
Fig. 4 is that a kind of dynamic image unsharp masking provided in an embodiment of the present invention goes the module of artifact intensifier to illustrate
Figure.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Due to the puppet produced when presence can not be removed and handled using USM enhancing algorithms dynamic image in the prior art
The technical problem of shadow.
In order to solve the above-mentioned technical problem, the present invention propose a kind of dynamic image unsharp masking go artifact Enhancement Method and
Device, as shown in figure 1, being decomposed using USM enhancing algorithms to the pending dynamic image of input, obtains high frequency detail figure
Picture and low frequency contour images；Generating details enhancing according to the high frequency detail image goes artifact mask and dynamic range compression to go puppet
Shadow mask；Artifact mask and dynamic range compression is gone to go artifact mask to enter the high frequency detail image using details enhancing
Row goes the enhancing of artifact details to handle, and obtains the enhanced high frequency detail image of artifact details, the low frequency contour images are entered
Mobile state Ratage Coutpressioit, the low frequency contour images after being compressed；Synthesis go the enhanced high frequency detail image of artifact details and
Low frequency contour images after compression, so as to obtain the dynamic image of pseudo- movie queen, and remove the dynamic image of pseudo- movie queen described in output.
A kind of dynamic image unsharp masking that the lower mask body introduction present invention is provided goes artifact Enhancement Method, such as Fig. 2 institutes
Show, including：
Step S1, is decomposed to pending dynamic image using unsharp masking USM enhancing algorithms, obtains high frequency thin
Save image and low frequency contour images；
Specifically, the method that the present invention is provided is mainly used for handling dynamic image, but is not limited only to dynamic
The processing of image.
Specifically, strengthen algorithm using USM and pending dynamic image is decomposed into high frequency detail image and low frequency profile
Image；Wherein, decompose and obtain the process of high frequency detail image and be：Set according to pending dynamic image Srclmg details size
Contiguous range, strengthens algorithm using USM and carries out mean filter Fuzzy Processing to the contiguous range of each pixel, obtain low frequency wheel
Wide image Lowlmg；Decompose and obtain the processes of low frequency contour images and be：The low frequency wheel is subtracted using pending dynamic image
Wide image Lowlmg, obtains high frequency detail image Highlmg.
Step S2, generates details enhancing according to the high frequency detail image and goes artifact mask and dynamic range compression to go artifact
Mask；
Specifically, step S2 is specifically included：
Step S21, carries out binary conversion treatment to the high frequency detail image, obtains negative edge region mask, the negative side
Edge region mask is：
Wherein, NegMask is negative edge region mask, and HighImg (r, c) is that coordinate is (r, c) in high frequency detail image
Point pixel value, wherein r represents row coordinate, and c represents row coordinate；
Step S22, the information in negative edge region is extracted according to the negative edge region mask, and to the negative edge region
Information be normalized, obtain the information in the negative edge region after normalized；
Wherein, the formula of the information in extraction negative edge region is：
NegHigh=NegMask*HighImg,
Wherein, NegHigh is the information in negative edge region；
Wherein, the formula of normalized is：
Wherein, MinVal=min (NegHigh), MaxVal=max (NegHigh).
Step S23, erosion operation is done with reference to the contiguous range of setting to the information in the negative edge region after normalized,
And expand negative edge regional extent, obtain expanding the information in the negative edge region of negative edge regional extent；
Specifically, when doing erosion operation to the information in negative edge region, corrosion neighborhood is that USM decomposes neighborhood model used
Enclose；Information to negative edge region is done after erosion operation, retains the information of non-negative fringe region, expands negative edge regional extent,
Obtain expanding the information in the negative edge region of negative edge regional extent.
Wherein, the formula of erosion operation is：
ErodeNegHigh=Erode (NegHigh, NeibSize),
Wherein, ErodeNegHigh is the information in the negative edge region after corrosion, and Erode () is to negative edge region
Information is corroded, and NeibSize is contiguous range set when USM is decomposed；
Wherein, the formula of expansion negative edge regional extent is：
NegHigh=NegMask*ErodeNegHigh+ (1-NegMask) * NegHigh；
Step S24, enters line translation stretching to expanding the information in the negative edge region of negative edge regional extent, is become
The information in the negative edge region changed after stretching；The formula for wherein converting stretching is：
NegHigh=NegHighgamma,
Wherein, gamma spans are 3~7；
Step S25, artifact mask is removed using the information generation in the negative edge region after obtained conversion stretching；
Wherein, generation goes the formula of artifact mask to be：
DeArtiMask=1-deArtiParam+deArtiParam*NegHigh
Wherein, deArtiParam is removes artifact mask, and deArtiParam is artifact strength, and span is 0.0~
1.0；
Step S26, goes the acquisition details enhancing of artifact mask to go artifact mask and dynamic range compression to go artifact according to described
Mask；
Wherein, obtaining details enhancing goes the formula of artifact mask to be：
EeDeArtiMask=1+eeParam*DeArtiMask,
Wherein, eeParam is that details strengthens intensive parameter, and EeDeArtiMask is that artifact mask is removed in details enhancing；
Wherein, obtaining dynamic range compression goes the formula of artifact mask to be：
DrcDeArtiMask=drcParam+ (1-drcParam) * DeArtiMask,
Wherein, drcParam is dynamic range compression parameter, and DrcDeArtiMask is that dynamic range compression goes artifact to cover
Mould.
Step S3, goes artifact mask and dynamic range compression to go artifact mask thin to the high frequency using details enhancing
Section image carries out going the enhancing of artifact details to handle, and the enhanced high frequency detail image of artifact details is obtained, to the low frequency wheel
Wide image carries out dynamic range compression, the low frequency contour images after being compressed；
Specifically, being multiplied by the details enhancing using high frequency detail image goes artifact mask and the dynamic range compression to go
Artifact mask, obtains the enhanced high frequency detail image of artifact details；Preset dynamic model is multiplied by using low frequency contour images
Compression parameters are enclosed, the low frequency contour images after being compressed.
Step S4, the low frequency contour images that synthesis is gone after the enhanced high frequency detail image of artifact details and compression, is obtained
Remove the dynamic image of pseudo- movie queen.
Specifically, by the low frequency contour images gone after the enhanced high frequency detail image of artifact details and the compression
It is added, obtains the dynamic image of pseudo- movie queen.
Specifically, step S3 and step S4 processing procedure can be formulated as：
EeDrcImg=drcParam*LowImg+EeDeArtiMask*DrcDeArtiMask*High Img,
Wherein, EeDrclmg is that details strengthens the final output image after dynamic range compression.
The dynamic image unsharp masking that the present invention is provided goes artifact Enhancement Method, causes artifact according to strong edge enhancing
Feature generation details enhancing goes artifact mask and dynamic range compression to remove artifact mask, and algorithm is being strengthened to pending using USM
Dynamic image when being handled, go artifact mask and dynamic range compression to remove artifact mask to height using the details enhancing of generation
Frequency detail pictures are handled so that high frequency detail image carries out going artifact to optimize while details enhancing is carried out, to eliminate
The artifact of generation, improves dynamic image display effect.
It is to strengthen algorithm using original USM to carry out the image after details enhancing is handled with dynamic range compression as shown in Figure 3 a
Effect, as shown in Figure 3 b for using technical scheme provided in an embodiment of the present invention carry out go artifact optimize after effect；By Fig. 3 a
Understand that Fig. 3 b go pseudo- movie queen to significantly improve display effect with Fig. 3 b contrasts.
A kind of dynamic image unsharp masking that the lower mask body introduction present invention is provided removes artifact intensifier, such as Fig. 4 institutes
Show, including：
Decomposing module 1, for being decomposed using unsharp masking USM enhancing algorithms to pending dynamic image, is obtained
To high frequency detail image and low frequency contour images；
Specifically, what the present invention was provided goes artifact intensifier to be mainly used for handling dynamic image, but not only
It is limited to the processing to dynamic image.
Specifically, the decomposing module 1 includes：Decompose low frequency contour images module 11 and decompose high frequency detail image module
12；
The decomposition low frequency contour images module 11, for according to pending dynamic image details size Srclmg settings
Contiguous range, strengthens algorithm using USM and carries out mean filter Fuzzy Processing to the contiguous range of each pixel, obtain low frequency wheel
Wide image Lowlmg；
The decomposition high frequency detail image module 12, for subtracting the low frequency profile diagram using pending dynamic image
As Lowlmg, high frequency detail image Highlmg is obtained.
Remove artifact mask generating module 2, for according to the high frequency detail image generate details enhancing go artifact mask and
Dynamic range compression removes artifact mask；
Specifically, it is described to go artifact mask generating module 2 to include：Negative edge region mask extraction module 21, negative edge area
Domain information extraction module 22, negative edge region extension module 23, conversion stretching module 24, the first mask generating module 25 and second
Mask generating module 26；
The negative edge region mask extraction module 21, for carrying out binary conversion treatment to the high frequency detail image, is obtained
To negative edge region mask, the negative edge region mask is：
Wherein, NegMask is negative edge region mask, and HighImg (r, c) is that coordinate is (r, c) in high frequency detail image
Point pixel value, wherein r represents row coordinate, and c represents row coordinate；
The negative edge area information extraction module 22, for extracting negative edge region according to the negative edge region mask
Information, and the information in the negative edge region is normalized, obtains the negative edge region after normalized
Information；
Wherein, the formula of the information in extraction negative edge region is：
NegHigh=NegMask*HighImg,
Wherein, NegHigh is the information in negative edge region；
Wherein, the formula of normalized is：
Wherein, MinVal=min (NegHigh), MaxVal=max (NegHigh).
Negative edge region extension module 23, for combining the contiguous range of setting to the negative edge after normalized
The information in region does erosion operation, and expands negative edge regional extent, obtains expanding the negative edge region of negative edge regional extent
Information；
Specifically, when doing erosion operation to the information in negative edge region, corrosion neighborhood is that USM decomposes neighborhood model used
Enclose；Information to negative edge region is done after erosion operation, retains the information of non-negative fringe region, expands negative edge regional extent,
Obtain expanding the information in the negative edge region of negative edge regional extent.
Wherein, the formula of erosion operation is：
ErodeNegHigh=Erode (NegHigh, NeibSize),
Wherein, ErodeNegHigh is the information in the negative edge region after corrosion, and Erode () is to negative edge region
Information is corroded, and NeibSize is contiguous range set when USM is decomposed；
Wherein, the formula of expansion negative edge regional extent is：
NegHigh=NegMask*ErodeNegHigh+ (1-NegMask) * NegHigh；
The conversion stretching module 24, the information for the negative edge region to expanding negative edge regional extent is carried out
Conversion stretching, obtains the information in the negative edge region after conversion stretching；
The formula for wherein converting stretching is：
NegHigh=NegHighgamma,
Wherein, gamma spans are 3~7；
First mask generating module 25, for the information generation using the negative edge region after obtained conversion stretching
Remove artifact mask；
Wherein, generation goes the formula of artifact mask to be：
DeArtiMask=1-deArtiParam+deArtiParam*NegHigh
Wherein, deArtiParam is removes artifact mask, and deArtiParam is artifact strength, and span is 0.0~
1.0；
Second mask generating module 26, for according to it is described go artifact mask obtain details strengthen go artifact mask and
Dynamic range compression removes artifact mask；
Wherein, obtaining details enhancing goes the formula of artifact mask to be：
EeDeArtiMask=1+eeParam*DeArtiMask,
Wherein, eeParam is that details strengthens intensive parameter, and EeDeArtiMask is that artifact mask is removed in details enhancing；
Wherein, obtaining dynamic range compression goes the formula of artifact mask to be：
DrcDeArtiMask=drcParam+ (1-drcParam) * DeArtiMask,
Wherein, drcParam is dynamic range compression parameter, and DrcDeArtiMask is that dynamic range compression goes artifact to cover
Mould.
Processing module 3, for going artifact mask and dynamic range compression to remove artifact mask to institute using details enhancing
State high frequency detail image to carry out going the enhancing of artifact details to handle, the enhanced high frequency detail image of artifact details is obtained, to institute
State low frequency contour images and carry out dynamic range compression, the low frequency contour images after being compressed；
Specifically, the processing module 3 includes：First processing module 31 and Second processing module 32；
The first processing module 31, artifact mask and institute are removed for being multiplied by the details enhancing using high frequency detail image
State dynamic range compression and remove artifact mask, obtain the enhanced high frequency detail image of artifact details；
The Second processing module 32, for being multiplied by preset dynamic range compression parameter using low frequency contour images, is obtained
Low frequency contour images after to compression.
Synthesis module 4, for synthesizing the low frequency profile diagram gone after the enhanced high frequency detail image of artifact details and compression
Picture, obtains the dynamic image of pseudo- movie queen.
Specifically, the synthesis module 4 is used to go the enhanced high frequency detail image of artifact details and the pressure by described
Low frequency contour images after contracting are added, and obtain the dynamic image of pseudo- movie queen.
Specifically, the processing module 3 and the processing procedure of synthesis module 4 can be formulated as：
EeDrcImg=drcParam*LowImg+EeDeArtiMask*DrcDeArtiMask*High Img,
Wherein, EeDrclmg is that details strengthens the final output image after dynamic range compression.
The dynamic image unsharp masking that the present invention is provided removes artifact intensifier, causes artifact according to strong edge enhancing
Feature generation details enhancing goes artifact mask and dynamic range compression to remove artifact mask, and algorithm is being strengthened to pending using USM
Dynamic image when being handled, go artifact mask and dynamic range compression to remove artifact mask to height using the details enhancing of generation
Frequency detail pictures are handled so that high frequency detail image carries out going artifact to optimize while details enhancing is carried out, to eliminate
The artifact of generation, improves dynamic image display effect.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
Any modifications, equivalent substitutions and improvements made within refreshing and principle etc., should be included in the scope of the protection.
Claims (10)
1. a kind of dynamic image unsharp masking goes artifact Enhancement Method, it is characterised in that including：
Pending dynamic image is decomposed using unsharp masking USM enhancing algorithms, high frequency detail image is obtained and low
Frequency contour images；
Generating details enhancing according to the high frequency detail image goes artifact mask and dynamic range compression to remove artifact mask；
Artifact mask and dynamic range compression is gone to go artifact mask to carry out the high frequency detail image using details enhancing
Go the enhancing of artifact details to handle, obtain the enhanced high frequency detail image of artifact details, the low frequency contour images are carried out
Dynamic range compression, the low frequency contour images after being compressed；
The low frequency contour images that synthesis is gone after the enhanced high frequency detail image of artifact details and compression, obtain the dynamic of pseudo- movie queen
State image.
2. unsharp masking as claimed in claim 1 goes artifact Enhancement Method, it is characterised in that according to the high frequency detail figure
Artifact mask and dynamic range compression is gone to remove artifact mask as generation details strengthens, including：
Binary conversion treatment is carried out to the high frequency detail image, negative edge region mask is obtained；
The information in negative edge region is extracted according to the negative edge region mask, and the information in the negative edge region is returned
One change is handled, and obtains the information in the negative edge region after normalized；
Erosion operation is done to the information in the negative edge region after normalized with reference to contiguous range set in advance, and expands negative
Fringe region scope, obtains expanding the information in the negative edge region of negative edge regional extent；
The information in the negative edge region to expanding negative edge regional extent enters line translation stretching, obtains negative after conversion stretching
The information of fringe region；
Artifact mask is removed using the information generation in the negative edge region after obtained conversion stretching；
The acquisition details enhancing of artifact mask is gone to go artifact mask and dynamic range compression to remove artifact mask according to described.
3. unsharp masking as claimed in claim 2 goes artifact Enhancement Method, it is characterised in that the negative edge region mask
For：
<mrow>
<mi>N</mi>
<mi>e</mi>
<mi>g</mi>
<mi>M</mi>
<mi>a</mi>
<mi>s</mi>
<mi>k</mi>
<mrow>
<mo>(</mo>
<mi>r</mi>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mo>{</mo>
<mrow>
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mrow>
<mi>H</mi>
<mi>i</mi>
<mi>g</mi>
<mi>h</mi>
<mi>Im</mi>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>r</mi>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mo><</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mi>H</mi>
<mi>i</mi>
<mi>g</mi>
<mi>h</mi>
<mi>Im</mi>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>r</mi>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mo>&GreaterEqual;</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>,</mo>
</mrow>
</mrow>
The formula for extracting the information in negative edge region is：
NegHigh=NegMask*HighImg,
The formula of normalized is：
<mrow>
<mi>N</mi>
<mi>e</mi>
<mi>g</mi>
<mi>H</mi>
<mi>i</mi>
<mi>g</mi>
<mi>h</mi>
<mo>=</mo>
<mfrac>
<mrow>
<mi>N</mi>
<mi>e</mi>
<mi>g</mi>
<mi>H</mi>
<mi>i</mi>
<mi>g</mi>
<mi>h</mi>
<mo>-</mo>
<mi>M</mi>
<mi>i</mi>
<mi>n</mi>
<mi>V</mi>
<mi>a</mi>
<mi>l</mi>
</mrow>
<mrow>
<mi>M</mi>
<mi>a</mi>
<mi>x</mi>
<mi>V</mi>
<mi>a</mi>
<mi>l</mi>
<mo>-</mo>
<mi>M</mi>
<mi>i</mi>
<mi>n</mi>
<mi>V</mi>
<mi>a</mi>
<mi>l</mi>
</mrow>
</mfrac>
<mo>,</mo>
</mrow>
The formula of erosion operation is：
ErodeNegHigh=Erode (NegHigh, NeibSize),
Expand negative edge regional extent formula be：
NegHigh=NegMask*ErodeNegHigh+ (1-NegMask) * NegHigh；
Converting the formula stretched is：
NegHigh=NegHighgamma,
Generation goes the formula of artifact mask to be：
DeArtiMask=1-deArtiParam+deArtiParam*NegHigh,
Obtaining details enhancing goes the formula of artifact mask to be：
EeDeArtiMask=1+eeParam*DeArtiMask,
Obtaining dynamic range compression goes the formula of artifact mask to be：
DrcDeArtiMask=drcParam+ (1-drcParam) * DeArtiMask,
Wherein, NegMask is negative edge region mask, and HighImg (r, c) is the point that coordinate is (r, c) in high frequency detail image
Pixel value, r represents row coordinate, and c represents row coordinate；NegHigh is the information in negative edge region；MinVal represents NegHigh
Minimum value, MaxVal represents NegHigh maximum；ErodeNegHigh is the information in the negative edge region after corrosion,
Erode () is that the information in negative edge region is corroded, and NeibSize represents contiguous range set in advance；Gamma values
Scope is 3~7；DeArtiMask is removes artifact mask, and deArtiParam is artifact strength, and span is 0.0~1.0；
EeParam is that details strengthens intensive parameter, and EeDeArtiMask is that artifact mask is removed in details enhancing；DrcParam is dynamic range
Compression parameters, DrcDeArtiMask is that dynamic range compression removes artifact mask.
4. unsharp masking as claimed in claim 3 goes artifact Enhancement Method, it is characterised in that described to be increased using the details
It is strong to go artifact mask and dynamic range compression to go artifact mask to carry out going the enhancing of artifact details to handle to the high frequency detail image,
The enhanced high frequency detail image of artifact details is obtained, including：
Being multiplied by the details enhancing using high frequency detail image goes artifact mask and the dynamic range compression to remove artifact mask, obtains
To removing the enhanced high frequency detail image of artifact details；
It is described to the low frequency contour images carry out dynamic range compression, the low frequency contour images after being compressed, including：
Preset dynamic range compression parameter is multiplied by using low frequency contour images, the low frequency contour images after being compressed.
5. unsharp masking as claimed in claim 4 goes artifact Enhancement Method, it is characterised in that artifact details is gone in the synthesis
Low frequency contour images after enhanced high frequency detail image and compression, obtain the dynamic image of pseudo- movie queen, including：
Go the enhanced high frequency detail image of artifact details to be added with the low frequency contour images after the compression by described, gone
The dynamic image of pseudo- movie queen.
6. a kind of dynamic image unsharp masking removes artifact intensifier, it is characterised in that including：
Decomposing module, for being decomposed using unsharp masking USM enhancing algorithms to pending dynamic image, obtains high frequency
Detail pictures and low frequency contour images；
Artifact mask generating module is removed, artifact mask and dynamic model are removed for generating details enhancing according to the high frequency detail image
Artifact mask is removed in confined pressure contracting；
Processing module, for going artifact mask and dynamic range compression to remove artifact mask to the high frequency using details enhancing
Detail pictures carry out going the enhancing of artifact details to handle, and the enhanced high frequency detail image of artifact details are obtained, to the low frequency
Contour images carry out dynamic range compression, the low frequency contour images after being compressed；
Synthesis module, for synthesizing the low frequency contour images gone after the enhanced high frequency detail image of artifact details and compression, is obtained
To the dynamic image for going pseudo- movie queen.
7. unsharp masking as claimed in claim 6 removes artifact intensifier, it is characterised in that described to go artifact mask to generate
Module is specifically included：Negative edge region mask extraction module, negative edge area information extraction module, negative edge region expand mould
Block, conversion stretching module, the first mask generating module and the second mask generating module；
The negative edge region mask extraction module, for carrying out binary conversion treatment to the high frequency detail image, obtains negative side
Edge region mask；
The negative edge area information extraction module, the letter for extracting negative edge region according to the negative edge region mask
Breath, and the information in the negative edge region is normalized, obtain the information in the negative edge region after normalized；
Negative edge region extension module, for combining contiguous range set in advance to the negative edge area after normalized
The information in domain does erosion operation, and expands negative edge regional extent, obtains expanding the negative edge region of negative edge regional extent
Information；
The conversion stretching module, the information for the negative edge region to expanding negative edge regional extent is entered line translation and drawn
Stretch, obtain the information in the negative edge region after conversion stretching；
First mask generating module, artifact is gone for the information generation using the negative edge region after obtained conversion stretching
Mask；
Second mask generating module, artifact mask and dynamic model are removed for going artifact mask to obtain details enhancing according to
Artifact mask is removed in confined pressure contracting.
8. unsharp masking as claimed in claim 7 removes artifact intensifier, it is characterised in that the negative edge region mask
For：
<mrow>
<mi>N</mi>
<mi>e</mi>
<mi>g</mi>
<mi>M</mi>
<mi>a</mi>
<mi>s</mi>
<mi>k</mi>
<mrow>
<mo>(</mo>
<mi>r</mi>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mo>{</mo>
<mrow>
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mrow>
<mi>H</mi>
<mi>i</mi>
<mi>g</mi>
<mi>h</mi>
<mi>Im</mi>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>r</mi>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mo><</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mi>H</mi>
<mi>i</mi>
<mi>g</mi>
<mi>h</mi>
<mi>Im</mi>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mi>r</mi>
<mo>,</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mo>&GreaterEqual;</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>,</mo>
</mrow>
</mrow>
The formula for extracting the information in negative edge region is：
NegHigh=NegMask*HighImg,
The formula of normalized is：
<mrow>
<mi>N</mi>
<mi>e</mi>
<mi>g</mi>
<mi>H</mi>
<mi>i</mi>
<mi>g</mi>
<mi>h</mi>
<mo>=</mo>
<mfrac>
<mrow>
<mi>N</mi>
<mi>e</mi>
<mi>g</mi>
<mi>H</mi>
<mi>i</mi>
<mi>g</mi>
<mi>h</mi>
<mo>-</mo>
<mi>M</mi>
<mi>i</mi>
<mi>n</mi>
<mi>V</mi>
<mi>a</mi>
<mi>l</mi>
</mrow>
<mrow>
<mi>M</mi>
<mi>a</mi>
<mi>x</mi>
<mi>V</mi>
<mi>a</mi>
<mi>l</mi>
<mo>-</mo>
<mi>M</mi>
<mi>i</mi>
<mi>n</mi>
<mi>V</mi>
<mi>a</mi>
<mi>l</mi>
</mrow>
</mfrac>
<mo>,</mo>
</mrow>
The formula of erosion operation is：
ErodeNegHigh=Erode (NegHigh, NeibSize),
Expand negative edge regional extent formula be：
NegHigh=NegMask*ErodeNegHigh+ (1-NegMask) * NegHigh；
Converting the formula stretched is：
NegHigh=NegHighgamma,
Generation goes the formula of artifact mask to be：
DeArtiMask=1-deArtiParam+deArtiParam*NegHigh,
Obtaining details enhancing goes the formula of artifact mask to be：
EeDeArtiMask=1+eeParam*DeArtiMask,
Obtaining dynamic range compression goes the formula of artifact mask to be：
DrcDeArtiMask=drcParam+ (1-drcParam) * DeArtiMask,
Wherein, NegMask is negative edge region mask, and HighImg (r, c) is the point that coordinate is (r, c) in high frequency detail image
Pixel value, r represents row coordinate, and c represents row coordinate；NegHigh is the information in negative edge region；MinVal represents NegHigh
Minimum value, MaxVal represents NegHigh maximum；ErodeNegHigh is the information in the negative edge region after corrosion,
Erode () is that the information in negative edge region is corroded, and NeibSize represents contiguous range set in advance；Gamma values
Scope is 3~7；DeArtiMask is removes artifact mask, and deArtiParam is artifact strength, and span is 0.0~1.0；
EeParam is that details strengthens intensive parameter, and EeDeArtiMask is that artifact mask is removed in details enhancing；DrcParam is dynamic range
Compression parameters, DrcDeArtiMask is that dynamic range compression removes artifact mask.
9. unsharp masking as claimed in claim 8 removes artifact intensifier, it is characterised in that the processing module includes：
First processing module and Second processing module；
The first processing module, artifact mask and the dynamic are removed for being multiplied by the details enhancing using high frequency detail image
Ratage Coutpressioit removes artifact mask, obtains the enhanced high frequency detail image of artifact details；
The Second processing module, for being multiplied by preset dynamic range compression parameter using low frequency contour images, is compressed
Low frequency contour images afterwards.
10. unsharp masking as claimed in claim 9 removes artifact intensifier, it is characterised in that the synthesis module, specifically
For going the enhanced high frequency detail image of artifact details to be added with the low frequency contour images after the compression by described, gone
The dynamic image of pseudo- movie queen.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710492125.4A CN107301627B (en) | 2017-06-26 | 2017-06-26 | Artifact removing and enhancing method and device for dynamic image unsharp mask |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710492125.4A CN107301627B (en) | 2017-06-26 | 2017-06-26 | Artifact removing and enhancing method and device for dynamic image unsharp mask |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107301627A true CN107301627A (en) | 2017-10-27 |
| CN107301627B CN107301627B (en) | 2020-07-17 |
Family
ID=60134991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710492125.4A Active CN107301627B (en) | 2017-06-26 | 2017-06-26 | Artifact removing and enhancing method and device for dynamic image unsharp mask |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107301627B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110070508A (en) * | 2019-04-23 | 2019-07-30 | 西安交通大学 | A kind of unsharp Enhancement Method based on threshold value and Linear Mapping |
| CN115409833A (en) * | 2022-10-28 | 2022-11-29 | 一道新能源科技(衢州)有限公司 | Hot spot defect detection method of photovoltaic panel based on unsharp mask algorithm |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101727658A (en) * | 2008-10-14 | 2010-06-09 | 深圳迈瑞生物医疗电子股份有限公司 | Image processing method and device |
| CN101980282A (en) * | 2010-10-21 | 2011-02-23 | 电子科技大学 | Infrared image dynamic detail enhancement method |
| CN104616255A (en) * | 2015-01-11 | 2015-05-13 | 北京工业大学 | Adaptive enhancement method based on mammographic image |
| CN106156754A (en) * | 2016-07-29 | 2016-11-23 | 浙江工业大学 | A kind of multi-modal preprocess method of finger based on maximum circumscribed matrix region of interesting extraction and bilateral filtering |
| CN106485680A (en) * | 2016-10-13 | 2017-03-08 | 上海联影医疗科技有限公司 | Method for correcting image and device |
| CN106780413A (en) * | 2016-11-30 | 2017-05-31 | 深圳市安健科技股份有限公司 | A kind of image enchancing method and device |
-
2017
- 2017-06-26 CN CN201710492125.4A patent/CN107301627B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101727658A (en) * | 2008-10-14 | 2010-06-09 | 深圳迈瑞生物医疗电子股份有限公司 | Image processing method and device |
| CN101980282A (en) * | 2010-10-21 | 2011-02-23 | 电子科技大学 | Infrared image dynamic detail enhancement method |
| CN104616255A (en) * | 2015-01-11 | 2015-05-13 | 北京工业大学 | Adaptive enhancement method based on mammographic image |
| CN106156754A (en) * | 2016-07-29 | 2016-11-23 | 浙江工业大学 | A kind of multi-modal preprocess method of finger based on maximum circumscribed matrix region of interesting extraction and bilateral filtering |
| CN106485680A (en) * | 2016-10-13 | 2017-03-08 | 上海联影医疗科技有限公司 | Method for correcting image and device |
| CN106780413A (en) * | 2016-11-30 | 2017-05-31 | 深圳市安健科技股份有限公司 | A kind of image enchancing method and device |
Non-Patent Citations (1)
| Title |
|---|
| SHUHANG WANG等: "Naturalness Preserved Enhancement Algorithm for Non-Uniform Illumination Images", 《IEEE TRANSACTION ON IMAGE PROCESSING》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110070508A (en) * | 2019-04-23 | 2019-07-30 | 西安交通大学 | A kind of unsharp Enhancement Method based on threshold value and Linear Mapping |
| CN115409833A (en) * | 2022-10-28 | 2022-11-29 | 一道新能源科技(衢州)有限公司 | Hot spot defect detection method of photovoltaic panel based on unsharp mask algorithm |
| CN115409833B (en) * | 2022-10-28 | 2023-01-31 | 一道新能源科技(衢州)有限公司 | Hot spot defect detection method of photovoltaic panel based on unsharp mask algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107301627B (en) | 2020-07-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2427462A1 (en) | Image processing device, image processing method, storage medium, and program | |
| CN107301627A (en) | Dynamic image unsharp masking removes artifact Enhancement Method and device | |
| CN104182939B (en) | Medical image detail enhancement method | |
| TWI433053B (en) | Method and system for image sharpness enhancement based on local feature of the image | |
| CN110211058A (en) | A kind of data enhancement methods of medical image | |
| CN101101669A (en) | Image reinforcement method for self-adaptive regulation according to edge and brightness | |
| CN104063848A (en) | Enhancement method and device for low-illumination image | |
| CN103093433A (en) | Natural image denoising method based on regionalism and dictionary learning | |
| CN112991199B (en) | Image high-low frequency decomposition noise removal method based on residual dense network | |
| CN106056565B (en) | A kind of MRI and PET image fusion method decomposed based on Multiscale Morphological bilateral filtering and contrast is compressed | |
| CN107292845A (en) | Based on the pyramidal dynamic image noise-reduction method of standard deviation and device | |
| CN110322404B (en) | Image enhancement method and system | |
| CN112598602A (en) | Mask-based method for removing Moire of deep learning video | |
| CN115496695A (en) | High-dynamic infrared image self-adaptive enhancement and compression method | |
| CN107169932A (en) | A kind of image recovery method based on Gauss Poisson mixed noise model suitable for neutron imaging system diagram picture | |
| JP2010045711A (en) | Moving image noise canceling device, method and program | |
| CN109658350A (en) | A kind of enhancing of night facial video image and noise-reduction method | |
| CN117252773A (en) | Image enhancement method and system based on self-adaptive color correction and guided filtering | |
| CN109064413B (en) | Image contrast enhancement method and image acquisition medical equipment adopting same | |
| CN106803236A (en) | Asymmetric correction method based on fuzzy field singular value decomposition | |
| CN107274372B (en) | Pyramid local contrast-based dynamic image enhancement method and device | |
| CN107392869B (en) | Face image filtering method based on edge-preserving filter | |
| CN116109743A (en) | Digital person generation method and system based on AI and image synthesis technology | |
| US8792747B2 (en) | Image processing method and device, and medical imaging instrument | |
| CN105160635A (en) | Image filtering method based on fractional order differential estimation gradient domain |
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 |