CN107169925A - The method for reconstructing of stepless zooming super-resolution image - Google Patents
The method for reconstructing of stepless zooming super-resolution image Download PDFInfo
- Publication number
- CN107169925A CN107169925A CN201710267753.2A CN201710267753A CN107169925A CN 107169925 A CN107169925 A CN 107169925A CN 201710267753 A CN201710267753 A CN 201710267753A CN 107169925 A CN107169925 A CN 107169925A
- Authority
- CN
- China
- Prior art keywords
- resolution
- image
- low
- super
- dictionary
- 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 68
- 238000012549 training Methods 0.000 claims abstract description 19
- 230000008569 process Effects 0.000 claims abstract description 11
- 238000001914 filtration Methods 0.000 claims description 18
- 230000009467 reduction Effects 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 7
- 230000015556 catabolic process Effects 0.000 claims description 4
- 238000006731 degradation reaction Methods 0.000 claims description 4
- 238000003708 edge detection Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 230000017105 transposition Effects 0.000 claims description 4
- 238000004321 preservation Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 7
- 238000013507 mapping Methods 0.000 abstract description 5
- 230000008859 change Effects 0.000 abstract description 3
- 238000012360 testing method Methods 0.000 abstract description 3
- 239000013598 vector Substances 0.000 description 5
- 238000000605 extraction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000012634 optical imaging Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000009394 selective breeding Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4053—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution
- G06T3/4076—Scaling of whole images or parts thereof, e.g. expanding or contracting based on super-resolution, i.e. the output image resolution being higher than the sensor resolution using the original low-resolution images to iteratively correct the high-resolution images
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Editing Of Facsimile Originals (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to a kind of method for reconstructing of stepless zooming super-resolution image, including：Dictionary training process, image reconstruction process, three process steps of dimensional variation process.The present invention proposes the method for setting up image library training dictionary, by image set off-line training dictionary library, obtains the mapping relations of high-definition picture and low-resolution image, optimizes image reconstruction step, method run time is effectively reduced, with adaptivity and high efficiency.Threshold value in the present invention in method is to optimize fixation after being based on a large amount of contrast tests, need not be changed during place, in the case where not influenceing image reconstruction effect, effectively raise image reconstruction efficiency.The super-resolution rebuilding image that the present invention is obtained can carry out the change of scale of arbitrary size.The demand in the case of difference can be met in engineer applied, there is preferable applicability.
Description
Technical field
The invention belongs to Image Reconstruction Technology field, and in particular to a kind of reconstruction side of stepless zooming super-resolution image
Method.
Background technology
Currently, image resolution ratio can be influenceed by factors such as imaging system, shooting environmentals, and obtained image resolution ratio is past
It is past to meet the demand of people.However, optical imaging system and its detector are limited to technical merit at present and processing is multiple again
Miscellaneous degree, it is impossible to improve system imaging quality from hardware point of view.Therefore, the appearance of image super-resolution rebuilding technology, has broken figure
As the limitation of resolution ratio.By way of software processing from a width or several low-resolution images degenerated, one is calculated
Panel height image in different resolution, obtains more abundant detailed information.Current super-resolution rebuilding technology is widely used in image procossing
And optical imaging field, obtain the attention of people.
At present, the rarefaction representation image super-resolution rebuilding algorithm based on study can effectively improve the resolution ratio of image,
And the yardstick that multiplying power is 3 is amplified to image to amplify.However, in certain special cases, the height obtained by this method
Image in different resolution does not reach required scale requirement.
Therefore, how to provide a kind of image rebuilding method allows the super-resolution rebuilding image obtained to carry out any chi
The conversion of degree, to reach that related industries requirement becomes most important.
The content of the invention
In order to solve the above-mentioned problems in the prior art, the invention provides a kind of stepless zooming super-resolution image
Method for reconstructing, comprise the following steps：
Step 1, Fuzzy Processing and N times of down-sampling are carried out according to degradation model to high-resolution sample image, obtains low point
Resolution sample image；
Step 2, using K-SVD methods, dictionary training is carried out using the low resolution sample image, high-resolution word is obtained
Allusion quotation and low-resolution dictionary；
Step 3, low-resolution image to be reconstructed pre-process obtaining several low-resolution image blocks；
Step 4, fixed threshold is calculated using the low resolution sample image；
Step 5, judge whether the pixel value of each low-resolution image block is less than the fixed threshold；If so,
Then judge the low-resolution image block as Poor information image block；If it is not, then judge the low-resolution image block as
High information quantity image block；
Step 6, the high information quantity image block is used to rebuild with rarefaction representation algorithm based on dictionary learning and obtains high letter
Breath amount rebuilds subregion；
Step 7, the Poor information image block using interpolation algorithm rebuild and obtain Poor information reconstruction subregion；
Step 8, the high information quantity subregion and the Poor information are rebuild into subregion progress image mosaic to be surpassed
Resolution reconstruction image；
Step 9, M times of interpolation of super-resolution rebuilding image progress is obtained by N*M (N=2,3,4 using interpolation algorithm；M
>0) super-resolution rebuilding image again.
In one embodiment of the invention, the step 2 comprises the following steps：
(21) characteristics of image of the low resolution sample image is extracted by feature extracting method, extraterrestrial target is obtained
High-resolution features information and low resolution characteristic information.
(22) the K-SVD methods are utilized, the high-resolution features information and the low resolution characteristic information are entered
Row joint training, obtains the high-resolution dictionary and the low-resolution dictionary.
In one embodiment of the invention, the step (22) comprises the following steps:
(221) low-resolution dictionary is trained using sparse K-SVD methods；
(222) high-resolution dictionary is calculated.
In one embodiment of the invention, the step 3 comprises the following steps：
(31) the first image is obtained to the low-resolution image denoising to be reconstructed；
(32) described first image deblurring is obtained into the second image；
(33) by second image is split according to fixed length and width, preservation is handled, and forms the low-resolution image
Block.
In one embodiment of the invention, the step 4 comprises the following steps：
(41) the low resolution sample image is subjected to block segmentation and obtains low resolution sample image block；
(42) marginal information of the low resolution sample image block is extracted using Boundary extracting algorithm, is counted each described low
The pixel value distribution situation of the pixel value of resolution ratio sample image block and all low resolution sample image blocks；
(43) according to the pixel value of each low resolution sample image block and all low resolution sample image blocks
Pixel value distribution situation choose X (X>1) individual candidate thresholds；
(44) each candidate thresholds are carried out using the rarefaction representation image super-resolution rebuilding algorithm based on study
Calculating forms multiple high-resolution candidate images；
(45) according to the operation time of the rarefaction representation image super-resolution rebuilding algorithm based on study and correspondence shape
Into the high-resolution candidate image resolution ratio, from X candidate thresholds select described information amount threshold value.
In one embodiment of the invention, the Boundary extracting algorithm is Canny operator edge detection algorithms.
In one embodiment of the invention, the step 6 comprises the following steps:
(61) image filtering processing is carried out for the high information quantity image block, and carries out high-frequency characteristic to extract acquisition high frequency
Information；
(62) KPCA dimensionality reductions are carried out to the high information quantity image block using the high-frequency information, realizes high dimensional data pressure
Contracting；
(63) data after being compressed using OMP algorithms to dimensionality reduction, which rebuild, obtains the high information quantity reconstruction subregion.
In one embodiment of the invention, described image filtering operation uses two-dimensional filtering operator filtering device group, described
Wave filter group is f={ f1,f2,f3,f4, wherein,
f1=[1, -1], f2=f1 T
f3=LOG, f3=f3 T
T operates for matrix transposition, and LOG is 5 × 5 two-dimensional filtering operator.
In one embodiment of the invention, the step (63) comprises the following steps：
(631) rarefaction representation coefficient β is calculated according to the OMP algorithms, wherein,
Wherein, y is the data after dimensionality reduction is decompressed, T0For given degree of rarefication, βiFor the daughter element in matrix β；
(632) the rarefaction representation coefficient β is multiplied with the high-resolution dictionary, obtains the high information quantity and rebuild son
Region.
In one embodiment of the invention, the interpolation algorithm is bicubic interpolation algorithm.
The embodiment of the present invention has the advantages that,
1st, the embodiment of the present invention proposes the method for setting up image library training dictionary, and image set off-line training dictionary library is obtained
To the mapping relations of high-definition picture and low-resolution image, optimize image reconstruction step, effectively reduce method run time,
With adaptivity and high efficiency.
2nd, the super-resolution rebuilding image that the present invention is obtained can carry out the change of scale of arbitrary size.In engineer applied
The demand in the case of difference can be met, there is preferable applicability.
3rd, the threshold value in the present invention in method is to optimize fixation after being based on a large amount of contrast tests, is not required to during place
It is changed, in the case where not influenceing image reconstruction effect, effectively raises image reconstruction efficiency.
4th, zoomed in and out again according to the image after super-resolution rebuilding, because algorithm is reconstructed high-frequency information, for scaling
Effect can be more preferable.
Brief description of the drawings
Fig. 1 is a kind of method for reconstructing schematic flow sheet of stepless zooming super-resolution image provided in an embodiment of the present invention；
Fig. 2 is a kind of dictionary training principle schematic provided in an embodiment of the present invention；
Fig. 3 is a kind of method for reconstructing principle schematic of stepless zooming super-resolution image provided in an embodiment of the present invention.
Embodiment
Further detailed description is done to the present invention with reference to specific embodiment, but embodiments of the present invention are not limited to
This.
Embodiment one
Fig. 1, Fig. 2 and Fig. 3 are referred to, Fig. 1 is a kind of stepless zooming super-resolution image provided in an embodiment of the present invention
Method for reconstructing schematic flow sheet；Fig. 2 is a kind of dictionary training principle schematic provided in an embodiment of the present invention；Fig. 3 is the present invention
A kind of method for reconstructing principle schematic for stepless zooming super-resolution image that embodiment is provided.The stepless zooming super-resolution figure
The method for reconstructing of picture, comprises the following steps：
Step 1, Fuzzy Processing and N times of down-sampling are carried out according to degradation model to high-resolution sample image, obtains low point
Resolution sample image；
Step 2, using K-SVD methods, dictionary training is carried out using the low resolution sample image, high-resolution word is obtained
Allusion quotation and low-resolution dictionary；
Wherein, the method that step 1 and step 2 set up image library training dictionary, image set off-line training dictionary library is obtained
The mapping relations of high-definition picture and low-resolution image, optimize image reconstruction step, effectively reduce method run time, tool
There are adaptivity and high efficiency.
Step 3, low-resolution image to be reconstructed pre-process obtaining several low-resolution image blocks；
Step 4, fixed threshold is calculated using the low resolution sample image；
Step 5, judge whether the pixel value of each low-resolution image block is less than the fixed threshold；If so,
Then judge the low-resolution image block as Poor information image block；If it is not, then judge the low-resolution image block as
High information quantity image block；
Wherein, step 5 significantly improves reconstruction speed while reconstruction quality is ensured, is adapted to time performance requirement
Higher application.
Step 6, the high information quantity image block is used to rebuild with rarefaction representation algorithm based on dictionary learning and obtains high letter
Breath amount rebuilds subregion；
Wherein, the rarefaction representation method for reconstructing based on study avoids the artificial selection of neighbour's number, rebuilds effect preferable.
Step 7, the Poor information image block using interpolation algorithm rebuild and obtain Poor information reconstruction subregion；
Step 8, the high information quantity subregion and the Poor information are rebuild into subregion progress image mosaic to be surpassed
Resolution reconstruction image；
Step 9, M times of interpolation of super-resolution rebuilding image progress is obtained by N*M (N=2,3,4 using interpolation algorithm；M
>0) super-resolution rebuilding image again.
Specifically, the step 2 comprises the following steps：
(21) characteristics of image of the low resolution sample image is extracted by feature extracting method, extraterrestrial target is obtained
High-resolution features information and low resolution characteristic information.
(22) the K-SVD methods are utilized, the high-resolution features information and the low resolution characteristic information are entered
Row joint training, obtains the high-resolution dictionary and the low-resolution dictionary.
Wherein, the step (22) comprises the following steps:
(221) low-resolution dictionary is trained using sparse K-SVD methods；
(222) high-resolution dictionary is calculated.
Specifically, the step 3 comprises the following steps：
(31) the first image is obtained to the low-resolution image denoising to be reconstructed；
(32) described first image deblurring is obtained into the second image；
(33) by second image is split according to fixed length and width, preservation is handled, and forms the low-resolution image
Block.
Specifically, the step 4 comprises the following steps：
(41) the low resolution sample image is subjected to block segmentation and obtains low resolution sample image block；
(42) marginal information of the low resolution sample image block is extracted using Boundary extracting algorithm, is counted each described low
The pixel value distribution situation of the pixel value of resolution ratio sample image block and all low resolution sample image blocks；
(43) according to the pixel value of each low resolution sample image block and all low resolution sample image blocks
Pixel value distribution situation choose X (X>1) individual candidate thresholds；
(44) each candidate thresholds are carried out using the rarefaction representation image super-resolution rebuilding algorithm based on study
Calculating forms multiple high-resolution candidate images；
(45) according to the operation time of the rarefaction representation image super-resolution rebuilding algorithm based on study and correspondence shape
Into the high-resolution candidate image resolution ratio, from X candidate thresholds select described information amount threshold value.
Preferably, the Boundary extracting algorithm is Canny operator edge detection algorithms；
Furthermore, the step 6 comprises the following steps:
(61) image filtering processing is carried out for the high information quantity image block, and carries out high-frequency characteristic to extract acquisition high frequency
Information；
(62) KPCA dimensionality reductions are carried out to the high information quantity image block using the high-frequency information, realizes high dimensional data pressure
Contracting；
(63) data after being compressed using OMP algorithms to dimensionality reduction, which rebuild, obtains the high information quantity reconstruction subregion.
Further, described image filtering operation uses two-dimensional filtering operator filtering device group, and the wave filter group is f=
{f1,f2,f3,f4, wherein,
f1=[1, -1], f2=f1 T
f3=LOG, f3=f3 T
T operates for matrix transposition, and LOG is 5 × 5 two-dimensional filtering operator.
Further, the step (63) comprises the following steps：
(631) rarefaction representation coefficient β is calculated according to the OMP algorithms, wherein,
Wherein, y is the data after dimensionality reduction is decompressed, T0For given degree of rarefication, βiFor the daughter element in matrix β.
(632) the rarefaction representation coefficient β is multiplied with the high-resolution dictionary, obtains the high information quantity and rebuild son
Region.
Preferably, described interpolation algorithm is bicubic interpolation algorithm.Such as linear interpolation, closest member can also be used
Method, bilinear interpolation method etc..
The present embodiment has the following advantages that：
1st, the embodiment of the present invention proposes the method for setting up image library training dictionary, and image set off-line training dictionary library is obtained
To the mapping relations of high-definition picture and low-resolution image, optimize image reconstruction step, effectively reduce method run time,
With adaptivity and high efficiency.
2nd, the super-resolution rebuilding image that the present invention is obtained can carry out the change of scale of arbitrary size.In engineer applied
The demand in the case of difference can be met, there is preferable applicability.
3rd, the threshold value in the present invention in method is to optimize fixation after being based on a large amount of contrast tests, is not required to during place
It is changed, in the case where not influenceing image reconstruction effect, effectively raises image reconstruction efficiency.
Embodiment two
On the basis of above-described embodiment one, the present embodiment provides the reconstruction side of another stepless zooming super-resolution image
Method, includes dictionary training process, image reconstruction process, three process steps of dimensional variation process.Specifically include following steps：
S1:Dictionary process of reconstruction.
S11：Using a large amount of high-resolution sample images, high-definition picture is subjected to mould according to revised degradation model
Paste and N times of down-sampling, obtain corresponding low resolution sample image.
S12：The low resolution sample image obtained to step 1 extracts characteristics of image by feature extracting method, obtains sky
Between the high-low resolution characteristic information of target be XsAnd Ys。
S13：Using K-SVD methods, joint training is carried out to high-low resolution characteristic information, high-low resolution word is obtained
Allusion quotation.
S13a：Train low-resolution dictionary.Base dictionary Φ selected complete DCT dictionaries, was asked using sparse K-SVD methods
Solution：
Then low-resolution dictionary Dl=Φ W, W are an atom representing matrixs.Compared with parsing dictionary model, double sparse words
Allusion quotation model provides adaptivity by the modification to W.
S13b：Calculate high-resolution dictionary.Assuming that high-resolution-low-resolution image block is in high-resolution-low resolution
Rate dictionary has identical rarefaction representation coefficient A to lower, then can be calculated by minimizing the approximate error in below equation
High-resolution dictionary Dh：
Solved using pseudoinverse：
Dh=XsA+=XsAT(AAT)-1 (3)
Wherein, subscript "+" represents pseudoinverse.
S2 image reconstruction processes.
S21：Low-resolution image to be reconstructed is pre-processed, wherein mainly including image denoising, image deblurring
With the operation of sample piecemeal.Its process step is：
S21a：To low-resolution image denoising to be reconstructed；
S21b：The image deblurring obtained to step S21a；
S21c：The image obtained to step S21b carries out block division, entire image is split according to fixed length and width,
Preserve；
S22：Fixed threshold is calculated using the low resolution sample image obtained in step S11, specific process step is as follows：
S22a：The a large amount of low resolution sample images obtained in step S11 are subjected to block segmentation and obtain image block, image
The length and width of block are with step S21c；
S22b：Image block marginal information is extracted using Boundary extracting algorithm, the information content and all figures of each image block is counted
As the information content distribution situation of block；
S22c：Information content highest in image block is chosen, the pixel value for obtaining the image block is F1, takes f=F1/4, then f*
40%<=threshold value<=f*60%, take it is some in the range of this represent threshold value, for example, volume can take represents threshold value as follows：F*40%,
F*45%, f*50%, f*55%, f*60%, calculate the corresponding rarefaction representation Image Super-resolution based on study of each threshold point
The resolution ratio of image after the reconstruction time of rate algorithm for reconstructing and reconstruction, can judge according to subjective assessment and PSNR.Then root
According to user's demand from it is some represent threshold value in determine fixed threshold, such as user's demand more focuses on the time, and that just heightens ratio,
Time is just fast；If opposite user focuses on rebuilding effect, that just turns down ratio, and what is brought is exactly to rebuild effect preferably but the time
It is longer.For example, being computed rear f*50% best suits user's demand, then fixed threshold=f*50% is taken.
S22d：The low-resolution image block to be reconstructed that input step S21 is obtained, image is extracted using Boundary extracting algorithm
Block edge information, when the image block information content is no more than the fixed threshold that step S22c is determined, the image block is Poor information figure
As block；Otherwise, it is high information quantity image block.Preferably, the Boundary extracting algorithm is Canny operator edge detection algorithms.
S23：Image filtering operations are carried out for high information quantity image block, high-frequency characteristic extraction are carried out, using two-dimensional filtering
Operator filtering device group, wave filter group used is f={ f1,f2,f3,f4, it is by four different wave filter groups into respectively：
f1=[1, -1], f2=f1 T
(4)
f3=LOG, f3=f3 T
(5)
Wherein superscript T representing matrixs transposition is operated, and LOG represents the two-dimensional filtering operator of one kind 5 × 5.It is special by high frequency
Image block high-frequency information is obtained after levying extraction operation, with xlRepresent.
S24：KPCA dimensionality reductions are carried out to the image block that step S23 is obtained, realize that high dimensional data compresses.Dimensionality reduction step is as follows：
S24a：High dimensional data set is expressed as X={ x1,x2,x3,…,xM},xi∈RD, KPCA methods are by non-linear
Mapping function x → Φ (x) ∈ F, wherein F is feature space, so just each data x can be mapped into a high dimensional feature sky
Between.
S24b：Kernel function will be carried out by Φ in point x to F respective operations, and thus obtained F data satisfaction
The condition of the heart, i.e.,：
Then the covariance matrix in feature space F is：
S24c：Seek c eigenvalue λ >=0 and characteristic vector
V ∈ F { 0 }, Cv=λ v (8)
Then have
(Φ(xv) Cv)=λ (Φ (xv)·v) (9)
Φ (x are represented by view of all characteristic vectors1),Φ(x2),…,Φ(xM) linear combination, i.e.,：
Then have：
In formula, v=1,2,3 ..., M define M × M dimension matrix Ksμv
Kμv:=(Φ (xμ)·Φ(xv)) (12)
S24d：Solve above formula and obtain characteristic value and characteristic vector, for data acquisition system in characteristic vector space VkProjection
It can be write as：
So, data are projected to the characteristic vector V of covariance matrixkOn, projection result (the namely table of low-dimensional data
Showing can y) be expressed as：
S25:Rarefaction representation image super-resolution weight based on study is carried out to the data after the compression of step S24 dimensionality reductions
Build, comprise the following steps that：
S25a：The low-dimensional data y obtained using OMP algorithms to step S24d is in low-resolution dictionary DlUnder rarefaction representation
Factor beta, that is, solve equation below：
Wherein T0For given degree of rarefication, βiFor the daughter element in matrix β.
S25b：By the rarefaction representation coefficient β tried to achieve and high-resolution dictionary DhIt is multiplied, the high information quantity rebuild is rebuild
Subregion, i.e.,：
X=Dhβ
Wherein X is the super-resolution subregion tried to achieve.
S26：The Poor information image block that step 22d is obtained using bicubic interpolation algorithm rebuild obtaining low information
Amount rebuilds subregion；
S27：High information quantity subregion and the Poor information are rebuild into subregion progress image mosaic and obtain super-resolution
Reconstruction image；
S28：Using bicubic interpolation algorithm by the super-resolution rebuilding image progress M times of interpolation obtain N*M (N=2,
3,4；M>0) super-resolution rebuilding image again.
Above content is to combine specific preferred embodiment further description made for the present invention, it is impossible to assert
The specific implementation of the present invention is confined to these explanations.For general technical staff of the technical field of the invention,
On the premise of not departing from present inventive concept, can also make it is some it is simple deduction or or replace, should all be considered as belonging to the present invention
Protection domain.
Claims (10)
1. a kind of method for reconstructing of stepless zooming super-resolution image, it is characterised in that comprise the following steps：
Step 1, Fuzzy Processing and N times of down-sampling are carried out according to degradation model to high-resolution sample image, obtains low resolution
Sample image；
Step 2, using K-SVD methods, using the low resolution sample image carry out dictionary training, obtain high-resolution dictionary and
Low-resolution dictionary；
Step 3, low-resolution image to be reconstructed pre-process obtaining several low-resolution image blocks；
Step 4, fixed threshold is calculated using the low resolution sample image；
Step 5, judge whether the pixel value of each low-resolution image block is less than the fixed threshold；If so, then sentencing
The fixed low-resolution image block is Poor information image block；If it is not, then judging that the low-resolution image block is believed as height
Breath amount image block；
Step 6, the high information quantity image block is used to rebuild with rarefaction representation algorithm based on dictionary learning and obtains high information quantity
Rebuild subregion；
Step 7, the Poor information image block using interpolation algorithm rebuild and obtain Poor information reconstruction subregion；
Step 8, the high information quantity subregion and the Poor information are rebuild into subregion progress image mosaic and obtains super-resolution
Rate reconstruction image；
Step 9, M times of interpolation of super-resolution rebuilding image progress is obtained by N*M (N=2,3,4 using interpolation algorithm；M>0)
Super-resolution rebuilding image again.
2. the method for reconstructing of stepless zooming super-resolution image according to claim 1, it is characterised in that the step 2
Comprise the following steps：
(21) characteristics of image of the low resolution sample image is extracted by feature extracting method, the high score of extraterrestrial target is obtained
Resolution characteristic information and low resolution characteristic information.
(22) the K-SVD methods are utilized, the high-resolution features information and the low resolution characteristic information are joined
Training is closed, the high-resolution dictionary and the low-resolution dictionary is obtained.
3. the method for reconstructing of stepless zooming super-resolution image according to claim 2, it is characterised in that the step
(22) comprise the following steps:
(221) low-resolution dictionary is trained using sparse K-SVD methods；
(222) high-resolution dictionary is calculated.
4. the method for reconstructing of stepless zooming super-resolution image according to claim 1, it is characterised in that the step 3
Comprise the following steps：
(31) the first image is obtained to the low-resolution image denoising to be reconstructed；
(32) described first image deblurring is obtained into the second image；
(33) by second image is split according to fixed length and width, preservation is handled, and forms the low-resolution image block.
5. the method for reconstructing of stepless zooming super-resolution image according to claim 1, it is characterised in that the step 4
Comprise the following steps：
(41) the low resolution sample image is subjected to block segmentation and obtains low resolution sample image block；
(42) marginal information of the low resolution sample image block is extracted using Boundary extracting algorithm, each low resolution is counted
The pixel value distribution situation of the pixel value of rate sample image block and all low resolution sample image blocks；
(43) according to the pixel value of each low resolution sample image block and the picture of all low resolution sample image blocks
Plain Distribution value situation chooses X (X>1) individual candidate thresholds；
(44) each candidate thresholds are calculated using the rarefaction representation image super-resolution rebuilding algorithm based on study
Form multiple high-resolution candidate images；
(45) according to operation time of the rarefaction representation image super-resolution rebuilding algorithm based on study and it is correspondingly formed
The resolution ratio of the high-resolution candidate image, selects described information amount threshold value from X candidate thresholds.
6. the method for reconstructing of stepless zooming super-resolution image according to claim 5, it is characterised in that the edge is carried
It is Canny operator edge detection algorithms to take algorithm.
7. the method for reconstructing of stepless zooming super-resolution image according to claim 1, it is characterised in that the step 6
Comprise the following steps:
(61) image filtering processing is carried out for the high information quantity image block, and carries out high-frequency characteristic to extract acquisition high frequency letter
Breath；
(62) KPCA dimensionality reductions are carried out to the high information quantity image block using the high-frequency information, realizes that high dimensional data compresses；
(63) data after being compressed using OMP algorithms to dimensionality reduction, which rebuild, obtains the high information quantity reconstruction subregion.
8. the method for reconstructing of stepless zooming super-resolution image according to claim 7, it is characterised in that described image is filtered
Ripple operation uses two-dimensional filtering operator filtering device group, and the wave filter group is f={ f1,f2,f3,f4, wherein,
f1=[1, -1], f2=f1 T
f3=LOG, f3=f3 T
T operates for matrix transposition, and LOG is 5 × 5 two-dimensional filtering operator.
9. the method for reconstructing of stepless zooming super-resolution image according to claim 7, it is characterised in that the step
(63) comprise the following steps：
(631) rarefaction representation coefficient β is calculated according to the OMP algorithms, wherein,
Wherein, y is the data after dimensionality reduction is decompressed, T0For given degree of rarefication, βiFor the daughter element in matrix β；
(632) the rarefaction representation coefficient β is multiplied with the high-resolution dictionary, obtains the high information quantity and rebuild sub-district
Domain.
10. the method for reconstructing of stepless zooming super-resolution image according to claim 1, it is characterised in that the interpolation
Algorithm is bicubic interpolation algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710267753.2A CN107169925B (en) | 2017-04-21 | 2017-04-21 | The method for reconstructing of stepless zooming super-resolution image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710267753.2A CN107169925B (en) | 2017-04-21 | 2017-04-21 | The method for reconstructing of stepless zooming super-resolution image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107169925A true CN107169925A (en) | 2017-09-15 |
| CN107169925B CN107169925B (en) | 2019-10-22 |
Family
ID=59812837
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710267753.2A Active CN107169925B (en) | 2017-04-21 | 2017-04-21 | The method for reconstructing of stepless zooming super-resolution image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107169925B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109064394A (en) * | 2018-06-11 | 2018-12-21 | 西安电子科技大学 | A kind of image super-resolution rebuilding method based on convolutional neural networks |
| CN111986078A (en) * | 2019-05-21 | 2020-11-24 | 四川大学 | Multi-scale core CT image fusion reconstruction method based on guide data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104463801A (en) * | 2014-12-04 | 2015-03-25 | 河西学院 | Multi-sensing-information fusion method based on self-adaptation dictionary learning |
| WO2015180055A1 (en) * | 2014-05-28 | 2015-12-03 | 北京大学深圳研究生院 | Super-resolution image reconstruction method and apparatus based on classified dictionary database |
| CN105427243A (en) * | 2015-10-29 | 2016-03-23 | 中国科学院自动化研究所 | Video super-resolution reconstruction method based on adaptive interpolation kernel learning |
| US9466132B2 (en) * | 2014-12-08 | 2016-10-11 | General Electric Company | Systems and methods for motion mitigation determinations |
| CN106408550A (en) * | 2016-09-22 | 2017-02-15 | 天津工业大学 | Improved self-adaptive multi-dictionary learning image super-resolution reconstruction method |
-
2017
- 2017-04-21 CN CN201710267753.2A patent/CN107169925B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015180055A1 (en) * | 2014-05-28 | 2015-12-03 | 北京大学深圳研究生院 | Super-resolution image reconstruction method and apparatus based on classified dictionary database |
| CN104463801A (en) * | 2014-12-04 | 2015-03-25 | 河西学院 | Multi-sensing-information fusion method based on self-adaptation dictionary learning |
| US9466132B2 (en) * | 2014-12-08 | 2016-10-11 | General Electric Company | Systems and methods for motion mitigation determinations |
| CN105427243A (en) * | 2015-10-29 | 2016-03-23 | 中国科学院自动化研究所 | Video super-resolution reconstruction method based on adaptive interpolation kernel learning |
| CN106408550A (en) * | 2016-09-22 | 2017-02-15 | 天津工业大学 | Improved self-adaptive multi-dictionary learning image super-resolution reconstruction method |
Non-Patent Citations (1)
| Title |
|---|
| 何阳 等: "稀疏阈值的超分辨率图像重建", 《中国光学》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109064394A (en) * | 2018-06-11 | 2018-12-21 | 西安电子科技大学 | A kind of image super-resolution rebuilding method based on convolutional neural networks |
| CN111986078A (en) * | 2019-05-21 | 2020-11-24 | 四川大学 | Multi-scale core CT image fusion reconstruction method based on guide data |
| CN111986078B (en) * | 2019-05-21 | 2023-02-10 | 四川大学 | Multi-scale core CT image fusion reconstruction method based on guide data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107169925B (en) | 2019-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106952228B (en) | Super-resolution reconstruction method of single image based on image non-local self-similarity | |
| Jin et al. | A flexible deep CNN framework for image restoration | |
| CN110969577B (en) | Video super-resolution reconstruction method based on deep double attention network | |
| Zhu et al. | Single image super-resolution using deformable patches | |
| Zhang et al. | Single image super-resolution with multiscale similarity learning | |
| Fan et al. | Image hallucination using neighbor embedding over visual primitive manifolds | |
| Ren et al. | Single image super-resolution using local geometric duality and non-local similarity | |
| US8538200B2 (en) | Systems and methods for resolution-invariant image representation | |
| CN104574336B (en) | Super-resolution image reconstruction system based on adaptive sub- mould dictionary selection | |
| CN108805808A (en) | A method of improving video resolution using convolutional neural networks | |
| CN102842115B (en) | Based on the compressed sensing image super-resolution rebuilding method of double dictionary study | |
| CN107993194B (en) | Super-resolution reconstruction method based on stationary wavelet transform | |
| CN104021523B (en) | A kind of method of the image super-resolution amplification based on marginal classification | |
| CN104657962B (en) | The Image Super-resolution Reconstruction method returned based on cascading linear | |
| CN103077511A (en) | Image super-resolution reconstruction method based on dictionary learning and structure similarity | |
| CN103150713A (en) | Image super-resolution method of utilizing image block classification sparse representation and self-adaptive aggregation | |
| CN106709945B (en) | A kind of quality evaluating method for super-resolution image | |
| CN105550989B (en) | The image super-resolution method returned based on non local Gaussian process | |
| Ren et al. | Adjusted non-local regression and directional smoothness for image restoration | |
| CN113781308A (en) | Image super-resolution reconstruction method and device, storage medium and electronic equipment | |
| CN106600533B (en) | Single image super resolution ratio reconstruction method | |
| CN109559278A (en) | Super resolution image reconstruction method and system based on multiple features study | |
| Chen et al. | Image denoising via deep network based on edge enhancement | |
| CN107169925B (en) | The method for reconstructing of stepless zooming super-resolution image | |
| CN108492252A (en) | Face image super-resolution reconstruction method based on secondary reconstruction |
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 |