CN106127712A - Image enchancing method and device - Google Patents
Image enchancing method and device Download PDFInfo
- Publication number
- CN106127712A CN106127712A CN201610510790.7A CN201610510790A CN106127712A CN 106127712 A CN106127712 A CN 106127712A CN 201610510790 A CN201610510790 A CN 201610510790A CN 106127712 A CN106127712 A CN 106127712A
- Authority
- CN
- China
- Prior art keywords
- layer
- image
- low
- frequency
- conversion
- 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
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000003384 imaging method Methods 0.000 claims abstract description 96
- 238000006243 chemical reaction Methods 0.000 claims abstract description 85
- 230000015556 catabolic process Effects 0.000 claims abstract description 30
- 230000009466 transformation Effects 0.000 claims description 68
- 238000000354 decomposition reaction Methods 0.000 claims description 40
- 230000008569 process Effects 0.000 claims description 10
- 238000003745 diagnosis Methods 0.000 description 17
- 230000008859 change Effects 0.000 description 10
- 230000002708 enhancing effect Effects 0.000 description 10
- 201000010099 disease Diseases 0.000 description 8
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 8
- 238000005070 sampling Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005728 strengthening Methods 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000002601 radiography 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20064—Wavelet transform [DWT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
A kind of image enchancing method and device, described image enchancing method includes: converting described picture breakdown based on first is L layer;Converting described picture breakdown based on second is L+N layer, N >=1;Decompose, based on the second conversion, the low-frequency image of L layer obtained and the first conversion is decomposed the ground floor that obtains to the high frequency imaging of L layer and is reconstructed.Technical solution of the present invention enhances the contrast of image to a great extent, improves the quality of image.
Description
Technical field
The present invention relates to technical field of image processing, particularly to a kind of image enchancing method and device.
Background technology
Digital X-ray photography (DR, Digital Radiography) equipment is computer digital image treatment technology and X
The armarium of a kind of advanced person that ray irradiation technology combines and formed.Digital X-ray photographic equipment because its radiation dose is little,
The accuracy of quality of image height, the recall rate of disease and diagnosis is higher and is widely used.
For digital X-ray photographic equipment, the contrast of the image that detector directly exports is relatively low, is unfavorable for doctor
Observation focus and the details that some are trickle.So that doctor can be convenient and diagnose the state of an illness of patient accurately, DR sets
Standby Image post-processing algorithm would generally use image enchancing method to increase the contrast of image so that the details of image is more
Add prominent and clear.At present, image DR equipment collected based on multiple dimensioned image enchancing method is generally used to carry out
Strengthen, such as methods such as wavelet transformation, Laplacian-pyramid image enhancings.But use current Enhancement Method to collecting
After image procossing, enhanced image of low quality, do not meet the clinical demand of reality, bring inconvenience to the diagnosis of doctor
While, it is also possible to cause failing to pinpoint a disease in diagnosis or the generation of Misdiagnosis.
Therefore, how to strengthen to improve picture quality to image, become one of current problem demanding prompt solution.
Summary of the invention
The problem to be solved in the present invention is to provide a kind of image enchancing method and device, to improve the contrast of image, enters
And improve the quality of image.
For solving the problems referred to above, technical solution of the present invention provides a kind of image enchancing method, including:
Converting described picture breakdown based on first is L layer;
Converting described picture breakdown based on second is L+N layer, N >=1;
Decompose, based on the second conversion, the low-frequency image of L layer obtained and the first conversion decomposes the ground floor that obtains to L
The high frequency imaging of layer is reconstructed.
Optionally, the described low-frequency image converting the L layer that decomposition obtains based on second and the first conversion decomposition obtain
Ground floor to the high frequency imaging of L layer is reconstructed and includes:
The low-frequency image decomposing the L layer obtained with the second conversion updates the first low frequency converting the L layer that decomposition obtains
Image;
The ground floor that obtains is decomposed to the high frequency imaging of L layer and the low-frequency image of the L layer after updating with the first conversion
It is reconstructed.
Optionally, described the ground floor that obtains is decomposed to the high frequency imaging of L layer and the L layer after updating with the first conversion
Low-frequency image be reconstructed and include:
The low-frequency image of L-i layer after updating is carried out bilinear interpolation or three interpolations to obtain the of L-i layer
One image, the high frequency imaging of the first image based on described L-i layer and L-i layer obtains the low of the L-i-1 layer after updating
Frequently image, 0≤i≤L-2;
Repeat said process, until obtaining the low-frequency image of the ground floor after updating;
It is reconstructed with the low-frequency image of the high frequency imaging of ground floor and the ground floor after updating.
Optionally, the described low-frequency image converting the L layer that decomposition obtains based on second and the first conversion decomposition obtain
Ground floor to the high frequency imaging of L layer is reconstructed and includes:
The ground floor obtaining the first conversion decomposition strengthens respectively to the high frequency imaging of L layer;
The low-frequency image decomposing the L layer obtained with the second conversion updates the first low frequency converting the L layer that decomposition obtains
Image;
It is reconstructed with the low-frequency image of the high frequency imaging of enhanced ground floor to L layer and the L layer after updating.
Optionally, described high frequency imaging with enhanced ground floor to L layer and the low-frequency image of the L layer after updating
It is reconstructed and includes:
The low-frequency image of L-i layer after updating is carried out bilinear interpolation or three interpolations to obtain the of L-i layer
Two images, the high frequency imaging of the second image based on described L-i layer and enhanced L-i layer obtains the L-i-after updating
The low-frequency image of 1 layer, 0≤i≤L-2;
Repeat said process, until obtaining the low-frequency image of the ground floor after updating;
It is reconstructed with the low-frequency image of the high frequency imaging of enhanced ground floor and the ground floor after updating.
Optionally, described second conversion decomposes the low-frequency image of the L layer obtained by decomposing described second conversion
To the low-frequency image of L+N layer and the L+1 layer that enhances to the high frequency imaging of L+N layer reconstruct and obtain.
Optionally, the low-frequency image L+N layer obtained being decomposed in the second conversion and the L+1 layer enhanced to L+N
The high frequency imaging reconstruct of layer includes:
The low-frequency image of L+N-i layer is carried out bilinear interpolation or three interpolations to obtain the 3rd figure of L+N-i layer
Picture, the high frequency imaging of the 3rd image based on described L+N-i layer and enhanced L+N-i layer reconstructs L+N-i-1 layer
Low-frequency image, 0≤i≤N-1;
Repeat said process, until the low-frequency image of reconstruct L layer.
Optionally, described first is transformed to Laplace transform, and second is transformed to wavelet transformation.
For solving the problems referred to above, technical solution of the present invention also provides for a kind of image intensifier device, including:
First resolving cell, is L layer for converting described picture breakdown based on first;
Second resolving cell, is L+N layer for converting described picture breakdown based on second, N >=1;
Reconfiguration unit, obtains for the low-frequency image and the first conversion decomposition decomposing the L layer obtained based on the second conversion
Ground floor to the high frequency imaging of L layer be reconstructed.
Optionally, described first resolving cell includes that Laplace transform module, described second resolving cell include small echo
Conversion module.
Compared with prior art, technical solution of the present invention has the advantage that
Converting described picture breakdown by first is L layer, and converting described picture breakdown by second is L+N layer, N >=
1 layer, and decompose the ground floor obtained to L layer based on the second low-frequency image converting the L layer that decomposition obtains and the first conversion
High frequency imaging be reconstructed, owing to the front L layer information that obtains being decomposed in the first conversion and the L+N obtained is decomposed in the second conversion
Layer information be effectively combined, and then both the small detail within image can be carried out enhancing can also be to image limit
Edge strengthens, and improves the quality of the image after reconstruct.For medical image, enhanced picture quality meets reality
Clinical demand, owing to the inside of image and edge are obtained for corresponding enhancing, and then reduce the most to a certain extent leakage
Examine or the probability of mistaken diagnosis.
Further, the second conversion decomposes the low-frequency image of the L layer obtained with by decomposing described second conversion
To the low-frequency image of L+N layer and the L+1 layer that enhances to the high frequency imaging of L+N layer reconstruct and obtain.Use which
Obtain the second conversion and decompose the low-frequency image of the L layer obtained, can further strengthen the right of the final image reconstructing and obtaining
Ratio degree, and then improve the quality of image.
Further, obtain at the low-frequency image and the first conversion decomposition decomposing the L layer obtained based on the second conversion
During ground floor to the high frequency imaging of L layer is reconstructed, the first conversion is decomposed the ground floor that obtains to the height of L layer
Frequently image strengthens, finally with the low-frequency image of L layer that have updated in the first conversion and the ground floor enhanced to L
The high frequency imaging of layer is reconstructed, and carries out again owing to the first conversion being decomposed the ground floor that obtains to the high frequency imaging of L layer
Strengthen, therefore, enhance the small detail within image and edge to a great extent, improve the matter of the image after reconstruct
Amount, for medical image, the picture quality after reconstruct meets the clinical demand of reality.
Further, when the low-frequency image that the second conversion is decomposed the L layer obtained is reconstructed, change L+1
Layer, to the interpolation method of the low-frequency image of L+N layer, can avoid occurring vibration artifact in the low-frequency image of L layer, with more
The first conversion after Xin is decomposed the ground floor that obtains to the low-frequency image of L layer and the first conversion and is decomposed the ground floor that obtains to the
When the high frequency imaging of L layer is reconstructed, the change interpolation method to the low-frequency image of each layer after updating, reconstruct can be removed
Vibration artifact present in the image obtained, owing to eliminating that may be present in the second conversion and the first conversion catabolic process shaking
Swing artifact, while improving picture contrast, therefore also been removed vibration artifact present in image, further increasing
The quality of image.For medical image, then the medical image finally obtained is made to more conform to the clinical demand of reality, more
Decrease well and fail to pinpoint a disease in diagnosis or the generation of Misdiagnosis.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the image enchancing method of embodiment of the present invention;
Fig. 2 is the structural representation of the image intensifier device of embodiment of the present invention.
Detailed description of the invention
Understandable, below in conjunction with the accompanying drawings to the present invention for enabling the above-mentioned purpose of the present invention, feature and advantage to become apparent from
Detailed description of the invention be described in detail.Elaborate detail in the following description so that fully understanding the present invention.But
Being that the present invention can be different from alternate manner described here implement with multiple, those skilled in the art can be without prejudice to this
Similar popularization is done in the case of invention intension.Therefore the present invention is not limited by following public detailed description of the invention.
As described in prior art, use current image enchancing method when image is strengthened, it is thus achieved that
Picture quality unsatisfactory, during as used wavelet transformation or Laplace transform that image is strengthened, enhanced figure
The contrast of picture is the most inconspicuous, and for medical image, enhanced image may not meet the clinical demand of reality, causes
Fail to pinpoint a disease in diagnosis or mistaken diagnosis.Inventor the analysis found that, when using wavelet transformation to carry out image enhaucament, the small detail of image can
Well to be strengthened, but the enhancing of the edge contour of image inconspicuous, and image is entered by this external employing wavelet transformation
When row strengthens, image there is also vibration artifact, therefore, uses wavelet transformation that image is strengthened, it is thus achieved that image
Of low quality.And when using Laplace transform to strengthen image, the edge contour of image can significantly be strengthened, but
It is the small detail reinforced partly of image inconspicuous, and when using Laplace transform to carry out image enhaucament, also has vibration
The existence of artifact, the quality of image is the highest.Therefore, inventor considers, if can be by wavelet transformation and Laplace transform
In conjunction with so that the final image obtained can significantly be strengthened on small detail and edge contour, i.e. strengthens image
Contrast;The vibration artifact additionally produced wavelet transformation and Laplace transform in image enhancement processes is carried out accordingly
Remove.
Refer to the schematic flow sheet that Fig. 1, Fig. 1 are the image enchancing methods of embodiment of the present invention, as it is shown in figure 1, institute
State image enchancing method to include:
S11: converting described picture breakdown based on first is L layer;
S12: converting described picture breakdown based on second is L+N layer, N >=1;
S13: decompose the ground floor obtained extremely based on the second low-frequency image converting the L layer that decomposition obtains and the first conversion
The high frequency imaging of L layer is reconstructed.
In present embodiment, after image is typically used the first conversion to decompose by the first conversion, the details letter of image
Breath be concentrated mainly on decompose with the first conversion obtain before in which floor high frequency imaging and low-frequency image, the second conversion is the most permissible
For to image use second conversion decompose after, the detailed information of image be concentrated mainly on second conversion decompose obtain rear which floor
In.As long as meeting the conversion of above-mentioned characteristic, all can convert as the first conversion or second, and based on above-mentioned Enhancement Method pair
Image strengthens.The reconstruct of image high frequency imaging based on the first ground floor of obtaining of conversion to L layer in present embodiment
Realize with the second low-frequency image converting the L layer obtained, both enhanced the small detail within image and also enhanced image
Edge.
Being transformed to Laplace transform with described first below, second is transformed to as a example by wavelet transformation embodiment party of the present invention
The image enchancing method of formula is described in detail.But the first conversion and the second conversion are not limited by technical scheme
It is fixed, as long as the conversion meeting features described above all can be as the first conversion or the second conversion.
Embodiment one
Perform S11: by Laplace transform, described image is carried out pyramid decomposition, obtain each layer of described image
High frequency imaging and low-frequency image.For example, to the employing Laplace transform of described image is decomposed into three layers, it is simply that by institute
State image and be first decomposed into the low-frequency image of ground floor and the high frequency imaging of ground floor, then the low-frequency image of ground floor is decomposed into
The low-frequency image of the second layer and the high frequency imaging of the second layer, then the low-frequency image of the second layer is decomposed into the low-frequency image of third layer
High frequency imaging with third layer.In the present embodiment, which floor is by Laplace transform by described picture breakdown, can be according to figure
Depending on actual demand in the actual size of picture and processing procedure, use Laplace transform that described image is entered in general
During row pyramid decomposition, which floor yardstick front is less, and the details of the image comprised is more, additionally in view of in actual process
Ageing, described picture breakdown can be three layers by Laplace transform by the present embodiment, namely L=3.
Perform S12: use wavelet transformation to be decomposed by described image, this step uses wavelet transformation described image is entered
The number of plies that row decomposes should be greater than the number of plies using Laplace transform to decompose described image.Still to use Laplce to become
As a example by described picture breakdown of changing commanders is three layers, then can use wavelet transformation is four layers by described picture breakdown, namely uses little
Described image is first decomposed into low-frequency image and the high frequency imaging of ground floor of ground floor by wave conversion, then by the low frequency of ground floor
Picture breakdown is low-frequency image and the high frequency imaging of the second layer of the second layer, and the low-frequency image of the second layer is decomposed into third layer
Low-frequency image and the high frequency imaging of third layer, be finally decomposed into the low-frequency image and the 4th of the 4th layer by the low-frequency image of third layer
The high frequency imaging of layer.Use wavelet transformation by the number of plies of described picture breakdown, ensureing more than using Laplace transform to figure
On the premise of carrying out the number of plies decomposed, it is also possible to depending on actual demand.
Performing S13, the low-frequency image of the L layer that employing wavelet transformation decomposition obtains and Laplace transform decomposition obtain
Ground floor is reconstructed to obtain enhanced image to the high frequency imaging of L layer.The present embodiment carries out weight in the following way
Structure is to obtain enhanced image.
From the foregoing, described image Laplace transform has been decomposed L layer by execution S11 and S12, (ground floor is extremely
L layer), decompose L+N layer (ground floor is to L+N layer) with wavelet transformation, in the present embodiment, obtained with wavelet transformation decomposition
The low-frequency image of L layer updates Laplace transform and decomposes the low-frequency image of the L layer obtained, namely decomposes with wavelet transformation
The low-frequency image of the L layer obtained is replaced Laplace transform and is decomposed the low-frequency image of the L layer obtained, and is becoming Laplce
Change and decompose after the low-frequency image of L layer obtained is updated, according to the low-frequency image of the L layer after updating come to its front the
L-1 layer is updated the most successively to the low-frequency image of ground floor, the most first updates the low-frequency image of L-1 layer, the most just
It is to add that the low-frequency image of the L layer after updating is to obtain the low frequency figure of the L-1 layer after updating with the high frequency imaging of L layer
Picture;Update the low-frequency image of L-2 layer again, specifically with the high frequency imaging of L-1 layer plus the L-1 layer after updating
Low-frequency image is to obtain the low-frequency image of the L-2 layer after updating;The like, until updating the low-frequency image to ground floor.
Laplace transform decomposition after the high frequency imaging of the ground floor then obtained with Laplace transform decomposition and renewal obtains
The low-frequency image of ground floor reconstructs enhanced image.
In the present embodiment, wavelet transformation decomposes the low-frequency image of the L layer obtained can be by decomposing wavelet transformation
The L+1 layer arrived is to the high frequency imaging of L+N layer, the L+1 layer obtained after carrying out linear or non-linear enhancing respectively to L
The high frequency imaging of+N shell, is then reconstructed acquisition by the low-frequency image that itself and wavelet transformation decompose the L+N layer obtained.Use
This mode obtains wavelet transformation and decomposes the low-frequency image of the L layer obtained, and can further strengthen what final reconstruct obtained
The contrast of image, improves the quality of image.
In other embodiments, the low-frequency image of the L layer that wavelet transformation decomposition obtains can also be directly by L+N layer
Low-frequency image and L+1 layer to the high frequency imaging of L+N layer reconstructs and obtains.
The number of plies decomposed image with employing Laplace transform below is three layers, uses wavelet transformation to enter image
After reconstructing based on Laplce and wavelet transformation in the present embodiment is strengthened as a example by being five layers by the number of plies that row decomposes with acquisition
The process of image carry out simple illustration.
Being three layers initially with Laplace transform by picture breakdown, using wavelet transformation is five layers by picture breakdown.So
After decompose, with wavelet transformation, the low-frequency image of third layer obtained and update and decompose the third layer that obtains with Laplace transform
Low-frequency image.
It is then to decompose, by wavelet transformation, the layer 5 obtained that wavelet transformation decomposes the low-frequency image of the third layer obtained
Low-frequency image, the high frequency imaging of the 4th layer and layer 5 has been carried out respectively non-linear enhanced 4th layer and layer 5
High frequency imaging obtains after being reconstructed.
After the low-frequency image obtaining the third layer that wavelet transformation decomposition obtains, decompose as Laplace transform
The low-frequency image of the third layer obtained, and on the basis of it, determine that the ground floor obtained is decomposed in the Laplace transform after renewal
With the low-frequency image of the second layer, concrete is carried out with the low-frequency image that Laplace transform is decomposed the third layer obtained exactly
The low-frequency image of the third layer after renewal decomposes the high frequency imaging of the third layer obtained to be updated plus Laplace transform
After the low-frequency image of the second layer, decompose plus Laplace transform with the low-frequency image of the second layer after the renewal obtained and obtain
The high frequency imaging of the second layer with the low-frequency image of the ground floor after being updated;Namely based on the low frequency of third layer after updating
The ground floor that image and employing Laplace transform decomposition obtain, to the high frequency imaging of third layer, recalculates Laplce successively
Conversion decomposes the ground floor that obtains to the low-frequency image of the second layer, and the low-frequency image and employing with the ground floor after updating draws general
Lars conversion is decomposed the high frequency imaging of the ground floor obtained and is reconstructed enhanced image.
In the present embodiment, Laplace transform is decomposed the front L layer information obtained and wavelet transformation decomposes the L+N obtained
The information of layer is effectively combined, and then strengthens the small detail within image and edge contour the most accordingly,
Improve picture quality.For medical image, enhanced picture quality meets the clinical demand of reality, also reduces leakage
Examine or the probability of mistaken diagnosis.
Embodiment two
In the present embodiment, image is carried out Laplace transform and is broken down into L layer (ground floor is to L layer), to image
Carrying out wavelet transformation and be broken down into L+N layer (ground floor is to L+N layer), similar with embodiment one, here is omitted.
The present embodiment is from the different of embodiment one, at the high frequency of the ground floor obtained with Laplace transform decomposition to L layer
Image and wavelet transformation decompose the low-frequency image of the L layer obtained when being reconstructed, in order to further strengthen the figure after reconstruct
The contrast of picture, the high frequency imaging of the ground floor that can obtain Laplace transform decomposition in the present embodiment to L layer enters respectively
Row strengthens, and specifically can linearly strengthen the high frequency imaging of ground floor to L layer or non-linear enhancing respectively.And draw
The renewal of the low-frequency image that the L layer obtained is decomposed in Laplace transform is similar with embodiment one, the most still divides with wavelet transformation
The low-frequency image of L layer that the L layer low-frequency image that solution obtains obtains to update Laplace transform to decompose, and based on updating after
The low-frequency image of L layer successively to its front L-1 layer, L-2 layer, L-3 layer ... the low-frequency image of ground floor is successively
Update.And the acquisition of the low-frequency image that wavelet transformation decomposes the L layer obtained is also similar with embodiment one, the most superfluous
State.
Finally obtain Laplace transform is decomposed after the ground floor that obtains has carried out strengthening to the high frequency imaging of L layer
Ground floor be reconstructed to L layer high frequency imaging and the low-frequency image of L layer after updating, to obtain enhanced figure
Picture.In the present embodiment, wavelet transformation decomposes the low-frequency image of the L layer obtained can decompose, by wavelet transformation, the L+ obtained
The L+1 layer that the low-frequency image of N shell and decompose wavelet transformation obtains obtains after strengthening to the high frequency imaging of L+N layer
Enhanced L+1 layer is reconstructed acquisition, and the low frequency figure of the L layer to obtain after reconstruct to the high frequency imaging of L+N layer
Be updated as the low-frequency image of L layer obtained is decomposed in Laplace transform, finally with in Laplace transform more
The low-frequency image of new L layer and the Laplace transform enhanced decompose the ground floor that obtains to the high frequency imaging of L layer
It is reconstructed, to a great extent the small detail within image and edge contour can be strengthened accordingly, improve
The quality of the image after reconstruct, and for medical image, the picture quality after reconstruct meets the clinical demand of reality, helps
The diagnosis of Yu doctor, also reduces and fails to pinpoint a disease in diagnosis or the probability of mistaken diagnosis.
Embodiment three
In the present embodiment, image is carried out Laplace transform and is broken down into L layer (ground floor is to L layer), to image
Carry out wavelet transformation and be broken down into L+N layer (ground floor is to L+N layer), decompose the L layer low frequency figure obtained with wavelet transformation
Decompose the low-frequency image of the L layer obtained as updating Laplace transform, and the low-frequency image with the L layer after updating is the most right
L-1 layer updates similar with embodiment one to the low-frequency image of ground floor accordingly, and here is omitted.This enforcement
Example is from embodiment one and the different of embodiment two, in view of using wavelet transformation and Laplace transform to divide in the present embodiment
When each tomographic image that solution obtains is reconstructed, has the existence of vibration artifact, the quality of image has been affected, therefore this enforcement
In example, the vibration artifact produced Laplace transform and wavelet transformation in restructuring procedure respectively eliminates accordingly, with
Under it is illustrated accordingly.
In the present embodiment, it is by obtaining wavelet transformation decomposition that wavelet transformation decomposes the low-frequency image of the L layer obtained
The low-frequency image of L+N layer and the L+1 layer that enhances to the high frequency imaging of L+N layer be reconstructed acquisition;This area
Skilled person will understand that employing wavelet transformation or Laplace transform are to need that image is carried out fall to adopt when decomposing image
Sample, therefore during reconstruct, need it is carried out a liter sampling, the most generally use the mode inserting 0 in low-frequency image
Carrying out a liter sampling, inventor finds through research, low-frequency image carries out in the way of inserting 0 liter sampling and can cause vibration puppet
The generation of shadow.Therefore, in order to eliminate the vibration artifact that wavelet transformation produces, in the mistake that the low-frequency image of L layer is reconstructed
Cheng Zhong, the low-frequency image that wavelet transformation decomposes the L+N layer obtained carries out liter sampling to obtain in the way of bilinear interpolation
The 3rd image of L+N layer, the L+N layer that the 3rd image based on described L+N layer and decompose wavelet transformation obtains
High frequency imaging carries out the high frequency imaging of enhanced L+N layer and decomposes the low frequency of the L+N-1 layer obtained with reconstruct wavelet transformation
Image;Then the low-frequency image to the L+N-1 layer after reconstruct continues to carry out liter sampling in the way of bilinear interpolation to obtain
Obtain the 3rd image of L+N-2 layer, the 3rd image based on described L+N-2 layer and the L+N-that wavelet transformation decomposition is obtained
The high frequency imaging of 2 layers carries out the high frequency imaging of enhanced L+N-2 layer and decomposes, with reconstruct wavelet transformation, the L+N-3 layer obtained
Low-frequency image, the rest may be inferred, until reconstruct wavelet transformation decomposes the low-frequency image of L layer obtained.Use bilinear interpolation
Mode carry out rise be sampled as state of the art, here is omitted.In other embodiments can also be respectively to L+1
Layer to the low-frequency image of L+N layer carries out three interpolations respectively and realizes adopting the liter of L+1 layer to the low-frequency image of L+N layer
Sample.
When the low-frequency image that wavelet transformation decomposes the L layer obtained is reconstructed, change L+1 layer to L+N
The interpolation method of low-frequency image of layer, can avoid occurring vibration artifact in the low-frequency image of L layer, and then after can also avoiding
When the continuous low-frequency image with L layer updates the low-frequency image that the L layer obtained is decomposed in Laplace transform, due to shaking that it exists
There is vibration artifact in the image that the final reconstruct swinging artifact and cause obtains.
Wavelet transformation is being decomposed after the vibration artifact in the low-frequency image of L layer obtained is removed, similarly, with
Wavelet transformation decomposes the low-frequency image of the L layer obtained and decomposes the low-frequency image of the L layer obtained to update Laplace transform
And front L-1 layer is to the low-frequency image of ground floor.In the present embodiment, it is contemplated that Laplace transform also likely to be present and shake
Swing artifact, therefore, decompose the low-frequency image of the L layer obtained with the Laplace transform after updating and to Laplace transform
Decompose the ground floor that obtains to the high frequency imaging of L layer to carry out enhanced ground floor to the high frequency imaging of L layer respectively and carry out
During reconstruct, still need to change the low-frequency image to L layer to ground floor and carry out the mode of interpolation to reduce the generation of vibration artifact.With
Sample ground, i.e. in restructuring procedure, the low-frequency image of L layer after updating is carried out in the way of bilinear interpolation liter sampling with
Obtaining the second image of L layer, the high frequency imaging of the second image based on described L layer and enhanced L layer obtains and updates
After the low-frequency image of L-1 layer, then proceed to the low-frequency image to the L-1 layer after updating and come in the way of bilinear interpolation
Carry out liter sampling to obtain the second image of L-1 layer, the second image based on described L-1 layer and enhanced L-1 layer
High frequency imaging obtain the low-frequency image of L-2 layer after updating, the rest may be inferred, until obtaining the low frequency of the ground floor after updating
Image, the Laplce after the high frequency imaging of the ground floor finally obtained with enhanced Laplace transform decomposition and renewal becomes
The low-frequency image changing the ground floor that decomposition obtains is reconstructed.The mode using bilinear interpolation carries out liter and is sampled as this area now
Having technology, here is omitted.In other embodiments can also be respectively to the ground floor after updating to the low-frequency image of L layer
Carry out three interpolations to realize the liter of the ground floor after updating to the low-frequency image of L layer is sampled.
Decompose at the low-frequency image and Laplace transform decomposing the L layer obtained with the Laplace transform after updating
To ground floor to the high frequency imaging of L layer be reconstructed time, change the interpolation side of low-frequency image to each layer after updating
Formula, can remove vibration artifact present in the image that reconstruct obtains, and the quality of the image improved further, for medical science figure
For Xiang, then make the medical image that finally obtains more conform to the clinical demand of reality, reduce and fail to pinpoint a disease in diagnosis or Misdiagnosis
Occur.
The number of plies decomposed image with employing Laplace transform below is three layers, uses wavelet transformation to enter image
After reconstructing based on Laplce and wavelet transformation in the present embodiment is strengthened as a example by being five layers by the number of plies that row decomposes with acquisition
The process of image carry out simple illustration.
Being three layers initially with Laplace transform by picture breakdown, using wavelet transformation is five layers by picture breakdown.So
After decompose, with wavelet transformation, the low-frequency image of third layer obtained and update and decompose the third layer that obtains with Laplace transform
Low-frequency image.
It is then by wavelet transformation is decomposed the 5th obtained that wavelet transformation decomposes the low-frequency image of the third layer obtained
The low-frequency image of layer carries out bilinear interpolation, with the low-frequency image of the layer 5 that carries out bilinear interpolation with carried out non-linear
The high frequency imaging of layer 5 strengthened reconstructs the low-frequency image obtaining the 4th layer, then with carry out bilinear interpolation the 4th
The low-frequency image of layer reconstructs what acquisition wavelet transformation decomposition obtained with the high frequency imaging of carried out non-linear enhancing the 4th layer
The low-frequency image of third layer.
After the low-frequency image obtaining the third layer that wavelet transformation decomposition obtains, decompose as Laplace transform
The low-frequency image of the third layer obtained, and on the basis of it, update Laplace transform and decompose the ground floor and the second layer obtained
Low-frequency image, the concrete low-frequency image that Laplace transform after updating is decomposed third layer exactly that obtain first carries out double
Linear interpolation, the low-frequency image of the third layer to carry out bilinear interpolation divides plus the non-linear Laplace transform enhanced
The high frequency imaging of the third layer that solution obtains is to obtain the low-frequency image of the second layer, then with the second layer carrying out bilinear interpolation
Low-frequency image decompose the high frequency imaging of the second layer obtained to obtain first plus the non-linear Laplace transform that enhances
The low-frequency image of layer, decomposes, plus the non-linear Laplace transform enhanced, the ground floor obtained with the low-frequency image of ground floor
High frequency imaging to reconstruct enhanced image.
In the present embodiment, Laplace transform is decomposed the front L layer information obtained and wavelet transformation decomposes the L+N obtained
The information of layer is effectively combined, and when obtaining the low-frequency image that wavelet transformation decomposes the L layer obtained, to be positioned at it
After the low-frequency image of each layer and the high frequency imaging enhanced it is reconstructed, then with the Laplace transform that have updated
Decompose the low-frequency image information of L layer obtained and the ground floor enhanced to the high frequency imaging of L layer to realize image reconstruction,
To a great extent the small detail within image and edge contour are strengthened.Additionally, wavelet transformation is being decomposed
To the low-frequency image of L layer be reconstructed during, change the mode of interpolation, dividing with the Laplace transform after updating
The Laplace transform of the low-frequency image of the L layer that solution obtains and enhancing decomposes the ground floor that obtains to the high frequency imaging of L layer
During being reconstructed, change the interpolation method to the ground floor after updating to the low-frequency image of L layer, and then make
The image obtained eventually had both had higher contrast and the most there is not vibration artifact, after improve reconstruct the most to a great extent
The quality of image, for medical image, the picture quality after reconstruct more conforms to the clinical demand of reality, contributes to doctor
Diagnosis while also reduce and fail to pinpoint a disease in diagnosis or the probability of mistaken diagnosis.
Corresponding to above-mentioned image enchancing method, embodiment of the present invention also provides for a kind of image intensifier device, refers to
Fig. 2, Fig. 2 are the structural representations of the image intensifier device of embodiment of the present invention, as in figure 2 it is shown, described image intensifier device
Including:
First resolving cell 10, is L layer for converting described picture breakdown based on first;
Second resolving cell 11, is L+N layer for converting described picture breakdown based on second, N >=1;
Reconfiguration unit 12, decomposes for the low-frequency image and the first conversion decomposing the L layer obtained based on the second conversion
To ground floor to the high frequency imaging of L layer be reconstructed.
In one embodiment, described first resolving cell includes Laplace transform module, described second resolving cell bag
Include wavelet transformation module.
Described image intensifier device be embodied as referring to the enforcement of described image enchancing method, do not repeat them here.
In sum, the image enchancing method that real mode of the present invention provides, at least have the advantages that
Converting described picture breakdown by first is L layer, and converting described picture breakdown by second is L+N layer, N >=
1 layer, and decompose the ground floor obtained to L layer based on the second low-frequency image converting the L layer that decomposition obtains and the first conversion
High frequency imaging be reconstructed, owing to the front L layer information that obtains being decomposed in the first conversion and the L+N obtained is decomposed in the second conversion
Layer information be effectively combined, and then both the small detail within image can be carried out enhancing can also be to image limit
Edge strengthens, and improves the quality of the image after reconstruct.For medical image, enhanced picture quality meets reality
Clinical demand, owing to the inside of image and edge are obtained for corresponding enhancing, and then reduce the most to a certain extent leakage
Examine or the probability of mistaken diagnosis.
Further, the second conversion decomposes the low-frequency image of the L layer obtained with by decomposing described second conversion
To the low-frequency image of L+N layer and the L+1 layer that enhances to the high frequency imaging of L+N layer reconstruct and obtain.Use which
Obtain the second conversion and decompose the low-frequency image of the L layer obtained, can further strengthen the right of the final image reconstructing and obtaining
Ratio degree, and then improve the quality of image.
Further, obtain at the low-frequency image and the first conversion decomposition decomposing the L layer obtained based on the second conversion
During ground floor to the high frequency imaging of L layer is reconstructed, the first conversion is decomposed the ground floor that obtains to the height of L layer
Frequently image strengthens, finally with the low-frequency image of L layer that have updated in the first conversion and the ground floor enhanced to L
The high frequency imaging of layer is reconstructed, and carries out again owing to the first conversion being decomposed the ground floor that obtains to the high frequency imaging of L layer
Strengthen, therefore, enhance the small detail within image and edge to a great extent, improve the matter of the image after reconstruct
Amount, for medical image, the picture quality after reconstruct meets the clinical demand of reality.
Further, when the low-frequency image that the second conversion is decomposed the L layer obtained is reconstructed, change L+1
Layer, to the interpolation method of the low-frequency image of L+N layer, can avoid occurring vibration artifact in the low-frequency image of L layer, with more
The first conversion after Xin is decomposed the ground floor that obtains to the low-frequency image of L layer and the first conversion and is decomposed the ground floor that obtains to the
When the high frequency imaging of L layer is reconstructed, the change interpolation method to the low-frequency image of each layer after updating, reconstruct can be removed
Vibration artifact present in the image obtained, owing to eliminating that may be present in the second conversion and the first conversion catabolic process shaking
Swing artifact, while improving picture contrast, therefore also been removed vibration artifact present in image, further increasing
The quality of image.For medical image, then the medical image finally obtained is made to more conform to the clinical demand of reality, more
Decrease well and fail to pinpoint a disease in diagnosis or the generation of Misdiagnosis.
Although the present invention is open as above with preferred embodiment, but it is not for limiting the present invention, any this area
Technical staff without departing from the spirit and scope of the present invention, may be by the method for the disclosure above and technology contents to this
Bright technical scheme makes possible variation and amendment, therefore, every content without departing from technical solution of the present invention, according to the present invention
Technical spirit any simple modification, equivalent variations and modification that above example is made, belong to technical solution of the present invention
Protection domain.
Claims (10)
1. an image enchancing method, it is characterised in that including:
Converting described picture breakdown based on first is L layer;
Converting described picture breakdown based on second is L+N layer, N >=1;
The ground floor obtained is decomposed to L layer based on the second low-frequency image converting the L layer that decomposition obtains and the first conversion
High frequency imaging is reconstructed.
2. image enchancing method as claimed in claim 1, it is characterised in that the described L obtained based on the second conversion decomposition
The low-frequency image of layer and the first conversion are decomposed the ground floor that obtains to the high frequency imaging of L layer and are reconstructed and include:
The low-frequency image decomposing the L layer obtained with the second conversion updates the first low-frequency image converting the L layer that decomposition obtains;
Decompose the ground floor that obtains to the low-frequency image of the high frequency imaging of L layer and the L layer after updating with the first conversion to carry out
Reconstruct.
3. image enchancing method as claimed in claim 2, it is characterised in that the described ground floor obtained with the first conversion decomposition
The low-frequency image of the L layer to the high frequency imaging and renewal of L layer is reconstructed and includes:
The low-frequency image of the L-i layer after updating is carried out bilinear interpolation or three interpolations to obtain the first figure of L-i layer
Picture, the high frequency imaging of the first image based on described L-i layer and L-i layer obtains the low frequency figure of the L-i-1 layer after updating
Picture, 0≤i≤L-2;
Repeat said process, until obtaining the low-frequency image of the ground floor after updating;
It is reconstructed with the low-frequency image of the high frequency imaging of ground floor and the ground floor after updating.
4. image enchancing method as claimed in claim 1, it is characterised in that the described L obtained based on the second conversion decomposition
The low-frequency image of layer and the first conversion are decomposed the ground floor that obtains to the high frequency imaging of L layer and are reconstructed and include:
The ground floor obtaining the first conversion decomposition strengthens respectively to the high frequency imaging of L layer;
The low-frequency image decomposing the L layer obtained with the second conversion updates the first low-frequency image converting the L layer that decomposition obtains;
It is reconstructed with the low-frequency image of the high frequency imaging of enhanced ground floor to L layer and the L layer after updating.
5. image enchancing method as claimed in claim 4, it is characterised in that described with enhanced ground floor to L layer
The low-frequency image of the L layer after high frequency imaging and renewal is reconstructed and includes:
The low-frequency image of the L-i layer after updating is carried out bilinear interpolation or three interpolations to obtain the second figure of L-i layer
Picture, the high frequency imaging of the second image based on described L-i layer and enhanced L-i layer obtains the L-i-1 layer after updating
Low-frequency image, 0≤i≤L-2;
Repeat said process, until obtaining the low-frequency image of the ground floor after updating;
It is reconstructed with the low-frequency image of the high frequency imaging of enhanced ground floor and the ground floor after updating.
6. the image enchancing method as described in any one of claim 2~5, it is characterised in that described second conversion decomposition obtains
The low-frequency image of L layer by the low-frequency image of L+N layer obtained and the L+1 enhanced are decomposed in described second conversion
Layer reconstructs acquisition to the high frequency imaging of L+N layer.
7. image enchancing method as claimed in claim 6, it is characterised in that to the L+N layer that the second conversion decomposition obtains
Low-frequency image includes with the high frequency imaging reconstruct of the L+1 layer enhanced to L+N layer:
The low-frequency image of L+N-i layer is carried out bilinear interpolation or three interpolations to obtain the 3rd image of L+N-i layer, base
The low frequency figure of L+N-i-1 layer is reconstructed in the 3rd image of described L+N-i layer and the high frequency imaging of enhanced L+N-i layer
Picture, 0≤i≤N-1;
Repeat said process, until the low-frequency image of reconstruct L layer.
8. image enchancing method as claimed in claim 1, it is characterised in that described first is transformed to Laplace transform, the
Two are transformed to wavelet transformation.
9. an image intensifier device, it is characterised in that including:
First resolving cell, is L layer for converting described picture breakdown based on first;
Second resolving cell, is L+N layer for converting described picture breakdown based on second, N >=1;
Reconfiguration unit, for decomposing, based on the second conversion, the low-frequency image of L layer obtained and the obtained is decomposed in the first conversion
One layer of high frequency imaging to L layer is reconstructed.
10. image intensifier device as claimed in claim 9, it is characterised in that described first resolving cell includes Laplce
Conversion module, described second resolving cell includes wavelet transformation module.
Priority Applications (14)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610510790.7A CN106127712B (en) | 2016-07-01 | 2016-07-01 | Image enhancement method and device |
RU2019134059A RU2797310C2 (en) | 2015-12-31 | 2016-12-29 | Methods and systems for image processing |
CN201680083009.0A CN108780571B (en) | 2015-12-31 | 2016-12-29 | Image processing method and system |
EP16881262.6A EP3398159B1 (en) | 2015-12-31 | 2016-12-29 | Methods and systems for image processing |
GB1710525.5A GB2548767B (en) | 2015-12-31 | 2016-12-29 | Methods and systems for image processing |
EP21174529.4A EP3920133A1 (en) | 2015-12-31 | 2016-12-29 | Methods and systems for image processing |
RU2018127759A RU2705014C1 (en) | 2015-12-31 | 2016-12-29 | Methods and systems for image processing |
BR112018013602-6A BR112018013602A2 (en) | 2015-12-31 | 2016-12-29 | image processing methods and systems |
PCT/CN2016/113079 WO2017114473A1 (en) | 2015-12-31 | 2016-12-29 | Methods and systems for image processing |
CN202210868915.9A CN115049563A (en) | 2015-12-31 | 2016-12-29 | Image processing method and system |
US15/638,327 US10290108B2 (en) | 2015-12-31 | 2017-06-29 | Methods and systems for image processing |
US16/410,119 US11049254B2 (en) | 2015-12-31 | 2019-05-13 | Methods and systems for image processing |
US17/361,315 US11880978B2 (en) | 2015-12-31 | 2021-06-28 | Methods and systems for image processing |
US18/419,528 US20240185425A1 (en) | 2015-12-31 | 2024-01-22 | Methods and systems for image processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610510790.7A CN106127712B (en) | 2016-07-01 | 2016-07-01 | Image enhancement method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106127712A true CN106127712A (en) | 2016-11-16 |
CN106127712B CN106127712B (en) | 2020-03-31 |
Family
ID=57467810
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610510790.7A Active CN106127712B (en) | 2015-12-31 | 2016-07-01 | Image enhancement method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106127712B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017114473A1 (en) * | 2015-12-31 | 2017-07-06 | Shanghai United Imaging Healthcare Co., Ltd. | Methods and systems for image processing |
CN108986074A (en) * | 2018-06-12 | 2018-12-11 | 中国矿业大学 | A kind of underground coal mine image processing method based on piecemeal Corresponding Sparse Algorithm |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101030298A (en) * | 2007-03-29 | 2007-09-05 | 杭州电子科技大学 | Method for enhancing medical image with multi-scale self-adaptive contrast change |
CN101109718A (en) * | 2006-11-14 | 2008-01-23 | 北京国药恒瑞美联信息技术有限公司 | Virtual grid imaging method and system used for eliminating influence of scattered radiation |
CN101201937A (en) * | 2007-09-18 | 2008-06-18 | 上海医疗器械厂有限公司 | Digital image enhancement method and device based on wavelet restruction and decompose |
CN101212560A (en) * | 2007-12-21 | 2008-07-02 | 上海广电集成电路有限公司 | Method for improving video image sharpness point by point |
CN101609549A (en) * | 2009-07-24 | 2009-12-23 | 河海大学常州校区 | The multi-scale geometric analysis super-resolution processing method of video blurred image |
CN101739671A (en) * | 2009-12-30 | 2010-06-16 | 中兴通讯股份有限公司 | Method and system for reducing noise of noisy images |
CN102324021A (en) * | 2011-09-05 | 2012-01-18 | 电子科技大学 | Infrared dim-small target detection method based on shear wave conversion |
CN102428478A (en) * | 2009-04-17 | 2012-04-25 | 里弗兰医疗集团公司 | Multi-scale image normalization and enhancement |
CN103049898A (en) * | 2013-01-27 | 2013-04-17 | 西安电子科技大学 | Method for fusing multispectral and full-color images with light cloud |
US20150193917A1 (en) * | 2014-01-09 | 2015-07-09 | Northrop Grumman Systems Corporation | Artificial vision system |
CN105931201A (en) * | 2016-04-20 | 2016-09-07 | 北京航空航天大学 | Image subjective visual effect enhancing method based on wavelet transformation |
-
2016
- 2016-07-01 CN CN201610510790.7A patent/CN106127712B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101109718A (en) * | 2006-11-14 | 2008-01-23 | 北京国药恒瑞美联信息技术有限公司 | Virtual grid imaging method and system used for eliminating influence of scattered radiation |
CN101030298A (en) * | 2007-03-29 | 2007-09-05 | 杭州电子科技大学 | Method for enhancing medical image with multi-scale self-adaptive contrast change |
CN101201937A (en) * | 2007-09-18 | 2008-06-18 | 上海医疗器械厂有限公司 | Digital image enhancement method and device based on wavelet restruction and decompose |
CN101212560A (en) * | 2007-12-21 | 2008-07-02 | 上海广电集成电路有限公司 | Method for improving video image sharpness point by point |
CN102428478A (en) * | 2009-04-17 | 2012-04-25 | 里弗兰医疗集团公司 | Multi-scale image normalization and enhancement |
CN101609549A (en) * | 2009-07-24 | 2009-12-23 | 河海大学常州校区 | The multi-scale geometric analysis super-resolution processing method of video blurred image |
CN101739671A (en) * | 2009-12-30 | 2010-06-16 | 中兴通讯股份有限公司 | Method and system for reducing noise of noisy images |
CN102324021A (en) * | 2011-09-05 | 2012-01-18 | 电子科技大学 | Infrared dim-small target detection method based on shear wave conversion |
CN103049898A (en) * | 2013-01-27 | 2013-04-17 | 西安电子科技大学 | Method for fusing multispectral and full-color images with light cloud |
US20150193917A1 (en) * | 2014-01-09 | 2015-07-09 | Northrop Grumman Systems Corporation | Artificial vision system |
CN105931201A (en) * | 2016-04-20 | 2016-09-07 | 北京航空航天大学 | Image subjective visual effect enhancing method based on wavelet transformation |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017114473A1 (en) * | 2015-12-31 | 2017-07-06 | Shanghai United Imaging Healthcare Co., Ltd. | Methods and systems for image processing |
GB2548767A (en) * | 2015-12-31 | 2017-09-27 | Shanghai United Imaging Healthcare Co Ltd | Methods and systems for image processing |
GB2548767B (en) * | 2015-12-31 | 2018-06-13 | Shanghai United Imaging Healthcare Co Ltd | Methods and systems for image processing |
US10290108B2 (en) | 2015-12-31 | 2019-05-14 | Shanghai United Imaging Healthcare Co., Ltd. | Methods and systems for image processing |
US11049254B2 (en) | 2015-12-31 | 2021-06-29 | Shanghai United Imaging Healthcare Co., Ltd. | Methods and systems for image processing |
US11880978B2 (en) | 2015-12-31 | 2024-01-23 | Shanghai United Imaging Healthcare Co., Ltd. | Methods and systems for image processing |
CN108986074A (en) * | 2018-06-12 | 2018-12-11 | 中国矿业大学 | A kind of underground coal mine image processing method based on piecemeal Corresponding Sparse Algorithm |
CN108986074B (en) * | 2018-06-12 | 2022-05-17 | 中国矿业大学 | Coal mine underground image processing method based on blocking sparse algorithm |
Also Published As
Publication number | Publication date |
---|---|
CN106127712B (en) | 2020-03-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111325686B (en) | Low-dose PET three-dimensional reconstruction method based on deep learning | |
CN109785243B (en) | Denoising method and computer based on unregistered low-dose CT of countermeasure generation network | |
JP4919408B2 (en) | Radiation image processing method, apparatus, and program | |
CN107958471B (en) | CT imaging method and device based on undersampled data, CT equipment and storage medium | |
CN111540025B (en) | Predicting images for image processing | |
CN110223255B (en) | Low-dose CT image denoising and recursion method based on residual error coding and decoding network | |
CN115953494B (en) | Multi-task high-quality CT image reconstruction method based on low dose and super resolution | |
CN102024251A (en) | System and method for multi-image based virtual non-contrast image enhancement for dual source CT | |
Podgorsak et al. | CT artifact correction for sparse and truncated projection data using generative adversarial networks | |
CN111968108B (en) | CT intelligent imaging method, device and system based on intelligent scanning protocol | |
CN108038840B (en) | Image processing method and device, image processing equipment and storage medium | |
CN113706409A (en) | CBCT image enhancement method and device based on artificial intelligence and storage medium | |
CN106228520B (en) | Image enchancing method and device | |
CN106127712A (en) | Image enchancing method and device | |
WO2023000244A1 (en) | Image processing method and system, and application of image processing method | |
Wang et al. | A deep network for sinogram and CT image reconstruction | |
CN111462004B (en) | Image enhancement method and device, computer equipment and storage medium | |
Li et al. | MARGANVAC: metal artifact reduction method based on generative adversarial network with variable constraints | |
CN113592968B (en) | Method and device for reducing metal artifacts in tomographic images | |
CN105125228A (en) | Image processing method for chest X-ray DR (digital radiography) image rib inhibition | |
CN113053496B (en) | Deep learning method for low-dose estimation of medical image | |
CN112446840B (en) | CT image black band artifact eliminating method and system based on deep learning | |
WO2021189383A1 (en) | Training and generation methods for generating high-energy ct image model, device, and storage medium | |
CN114305469A (en) | Low-dose digital breast tomography method and device and breast imaging equipment | |
Ruan et al. | QS-ADN: Quasi-supervised artifact disentanglement network for low-dose CT image denoising by local similarity among unpaired data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CP01 | Change in the name or title of a patent holder |
Address after: 201807 Shanghai city Jiading District Industrial Zone Jiading Road No. 2258 Patentee after: Shanghai Lianying Medical Technology Co., Ltd Address before: 201807 Shanghai city Jiading District Industrial Zone Jiading Road No. 2258 Patentee before: SHANGHAI UNITED IMAGING HEALTHCARE Co.,Ltd. |
|
CP01 | Change in the name or title of a patent holder |