CN106815876A - Image sparse characterizes the combined optimization training method of many dictionary learnings - Google Patents
Image sparse characterizes the combined optimization training method of many dictionary learnings Download PDFInfo
- Publication number
- CN106815876A CN106815876A CN201611252617.8A CN201611252617A CN106815876A CN 106815876 A CN106815876 A CN 106815876A CN 201611252617 A CN201611252617 A CN 201611252617A CN 106815876 A CN106815876 A CN 106815876A
- Authority
- CN
- China
- Prior art keywords
- dictionary
- image
- prime
- matrix
- represent
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
- G06T9/005—Statistical coding, e.g. Huffman, run length coding
Abstract
Image sparse characterizes many dictionary learning combined optimization training methods, belong to multimedia communication and image real time transfer field, it is characterized in that, the gradient matrix of training elementary area is regarded as through the nonzero element in the singular value matrix after singular value decomposition the energy value of correspondence gradient direction, elementary area is divided into isotropic image and anisotropic image by the energy value Parameters threshold according to setting, learn shared dictionary and specialized dictionary successively, with a reflection through the isotropism after sparse representation and the residual error of anisotropy image, the object function that the auto-correlation and the factor such as cross-correlation degree and nonzero element regularization of each dictionary are minimized is optimized, in optimization process, optimize A with orthogonal matching pursuit algorithm successively0,Ak, then optimize D with gradient descent algorithm0,Dk, when desire Optimal Parameters are retained, other items for not being related to be intended to Optimal Parameters are considered as constant.When the present invention is for compression of images, details is retained, and distortion rate is relatively low, and image quality is relatively preferable.
Description
Technical field
The invention provides a kind of image data compression method, belong to multimedia communication and data compression crossing domain, it is special
A kind of image data compression algorithm for low bit- rate is not designed, and cluster and the modeling of structuring dictionary are carried out to image texture, will
Image carries out sparse representation, is mainly used in reducing the data volume transmitted during communication, is not only suitable for the image such as face of particular topic,
It is applied to general natural image again, is widely used.
Background technology
Digital multimedia communications are most one of challenge, field with fastest developing speed in present communications technology various fields.
Compression and transmission of the big data epoch to data propose demand higher.In order to effectively mitigate bandwidth pressure, data are carried out
Effectively transmission, compression of images is widely studied by researchers.
Traditional method for compressing image can not produce good compression effectiveness in low bit- rate to image, in low bit- rate,
Recover image to be difficult to produce preferable visual effect.Such as the JPEG image compression method based on discrete cosine transform (DCT) is in weight
Obvious blocking effect can be produced when building, image block carries out transition coding, with the reduction of code check, be occurred in that not on the border of block
Continuously.The wavelet coefficient of high frequency has been carried out threshold value contraction by the JPEG2000 compression methods based on wavelet transform (DWT), is made
Into the loss of high-frequency information, ringing can be produced to image, its typical performance is that the field of gradation of image acute variation goes out
The concussion of Gibbs distribution is now similar to, the quality of restored image is had a strong impact on, and cause that successive image treatment is difficult to.
In recent years, dictionary learning method achieves preferable compression effectiveness, will data (image) it is sparse with one group of base
The base of linear combination represent that the dictionary is referred to " base ".It is identical with discrete cosine transform and wavelet transformation, dictionary
Habit is also to enter line translation according to base, and image is characterized in the transform domain as illustrated, in the hope of obtaining the expression effect more superior than pixel domain
Really, i.e., compression of images can be realized with less bit come phenogram picture in transform domain.Herein, corresponding to this group of base of dictionary
Coefficient is sparse, and major part is zero, only small part nonzero coefficient, and sparse volume is carried out to image using this superior property
Code (Sparse Coding, SC).Its difference is that discrete cosine base and wavelet basis are the bases obtained according to mathematical function,
And the base of dictionary learning, it is to be obtained with the Algorithm Learning of machine learning from real image.This have the advantage that, from reality
The border sample middle school more identical image of base out, and the base obtained according to mathematical function is difficult to the rule of perfect picture engraving.
The principle of dictionary learning is that the image study to training set obtains one group of redundancy base, and the redundancy base refers to the atom of dictionary
(i.e. a row of dictionary) number is more than its dimension, there is theoretical guarantee, and what the sample in test set can be with the redundancy base is several
Individual component linear combination uniquely represents, claims this to be expressed as sparse representation.Sparse representation is intended to minimum dictionary atom
A signal is represented, to realize representing image with a small number of data, the purpose of data compression is reached.Classical dictionary learning is calculated
Method has K-SVD, every time one atom of study renewal sparse coefficient corresponding with its, until all of atomic update is finished, repeats
Iteration can obtain complete dictionary several times.The deficiency of the algorithm is that the time of dictionary training is more long.Then, research carries
Go out online dictionary learning algorithm (Online Dictionary Learning), the method is reached by stochastic gradient descent algorithm
Optimal value, obtain dictionary by iteration several times, can quickly restrain.
Sparse coding is carried out to image, dictionary is generally divided into two kinds of forms of single dictionary and many dictionaries.Single dictionary format
It is that, to one unified dictionary of all image studies, this dictionary compacts, but lacks the resolution capability of different characteristic, it is difficult to
Optimal sign is carried out to all of feature.The representative algorithm that this unified dictionary is compressed sign to image has scholar
Recurrence least square dictionary learning algorithm (the Recursive Least Squares that Karl Skretting are proposed
Dictionary Learning Algorithm, RLS_DLA), the method is from pixel domain or wavelet field to different iconologies
Unified dictionary is practised out, low bit coding is carried out to image with sparse coefficient, algorithm iteration convergence is very fast.Many dictionary formats are
Feature different in image is modeled respectively, the dictionary of the multiple specializations of study, this dictionary has resolution capability.Scholar
Michael Elad propose to be characterized with the face image to people of multiple specialization dictionaries at first, i.e., to the feature of face such as
Eye, nose, eyebrow, mouth etc. set up the K-SVD dictionaries of specialization respectively, and each image block is independently characterized, this face characteristic word
Allusion quotation is compacted very much, realizes that low bit is encoded, and effect is better than coding standards such as traditional JPEG2000.But, the method adapts to energy
Power is limited, it is impossible to being compressed for other images in addition to face.Scholar Li Shaoyang is using the non-side for joining Bayesian learning
Natural image data are learnt multiple dictionaries by method, and the number of dictionary is automatically determined by algorithm according to prior information, this side
Method can learn to one group of optimal dictionary, and different images is characterized.
On the other hand, natural image has the general character of many inherences, and identical feature possibly be present in different images, or even
It is that most of images have.Image also has the characteristic of itself, and such as some image blocks have the directionality of distinctness, and geometric properties are bright
Really.Based on considerations above, it is necessary to set up a kind of new compression algorithm, the common feature to image sets up unified shared dictionary,
And multiple specialized dictionaries are set up to the property feature of image, and two kinds of mutual supplement with each other's advantages of the dictionary of form are realized, share dictionary
The sign that can be compacted to the realization of most of features, the dictionary of specialization can more added with resolving power, to shared dictionary and specialized word
Allusion quotation is optimized in combination, and study carries out sparse representation to one group of optimal dictionary to image.
The content of the invention
It is a kind of by learn shared dictionary that the general character of the isotropic images in image obtains and by learning image in
The proprietary dictionary that obtains of individual character carry out combined optimization training method, to realize that compression image is ensureing containing abundant information amount
While, compression ratio can be to greatest extent improved again.
It is an advantage of the current invention that different with multiple by the geometric component of modeled images different characteristic, i.e. low frequency component
The high fdrequency component of gradient direction so that the image rebuild after compression retains more complete details, there is more preferable Objective image quality,
And more meet the subjective feeling of people.
It is a feature of the present invention that being a kind of natural image coding method, learn optimal base carries out sparse table to image
Levy, realize according to the following steps successively in a computer:
Step (1), computer initialization sets following each parameters and coefficient:
Each image fritter will be obtained after training image X cuttingsI=1,2 ..., i ..., I, I are xiSum, xi
∈Rm, m is the row of matrix R, abbreviation xi,
Image fritter xiIn pixel useRepresent, write a Chinese character in simplified form into j, j=1,2 ..., J, meanwhile, j is also the sequence number of pixel, J
It is the sum of pixel, the horizontal component of the pixel is represented with u, vertical component is represented with v.
Use gjThe gradient of the pixel j is represented,Use GiRepresent described image fritter xi's
Gradient matrix, Gi=[g1,g2,…,gJ]T,Gi∈RJ×2, columns n=2 represent the number J of pixel gradient, row expression pixel with row
The level and vertical component of gradient, use respectivelyRepresent.
Deflections of the pixel j in gradient direction is represented with ω
Use K0Represent that the collection of isotropism subimage block is combined into, abbreviation K0, use D0Represent study K0The shared dictionary for obtaining.
The set of anisotropic image block, K={ K are represented with K1,K2,…,Kk,…,K6}={ Kk, referred to as 6 class gradients
The region k of deflection, wherein, K1Correspondence (0 °, 30 °), K2Correspondence (30 °+, 60 °), K3Correspondence (60 °+, 90 °), K4(90 ° of correspondence
+, 120 °), K5Correspondence (120 °+, 150 °), K6Correspondence (150 °+, 180 °), wherein, 30 °+represent that the rest may be inferred by analogy for it more than 30 °, symbol
Number " { } " expression " contains " element, similarly hereinafter.
Use DKRepresent the proprietary dictionary obtained after study K, abbreviation DK, correspondingly, use D1~D6Corresponding study is represented respectively
The 6 proprietary sub- dictionaries obtained after 6 anisotropy image blocks of the gradient angular zone, and DKBy 6 proprietary sub- dictionary groups
Into DK={ D1,D2,…,D6}={ Dk}。
The shared dictionary D is represented with D0With proprietary sub- dictionary DkSet, be expressed as:D={ D0, DK}={ D0,D1,D2,
D3,D4,D5,D6}。
With k=1,2 ..., 6 represent the subscript sequence of each proprietary sub- dictionaries.Dictionary D is represented with k '=0,1,2,3,4,5,6
In the contained sub- dictionary of whole subscript sequence.
Dictionary encoding coefficient matrix, A={ A are represented with A0,AK, wherein:A0Represent the shared dictionary D0Code coefficient
Matrix, AKRepresent the proprietary dictionary DKCode coefficient matrix, wherein,Represent AKIncluding:Proprietary son
Dictionary set { DkIn correspond to the set of anisotropy subimage block in each general character image part dictionary code coefficientAnd 6 class anisotropy subgraph each proprietary sub- dictionary DkCode coefficient matrixK=1,2 ..., 6.
The element of each code coefficient matrix is referred to as code coefficient a, δ<A < < 1, set code coefficient as combined optimization
One of parameter of training, its object is to:In follow-up study, when judge first through step (2.3) in target obtain, afterwards again pass through
When the compression image of parametric variable combined optimization is unsatisfactory for error requirements with the original input image of reality on image similarity, energy
By improving the similarity of compression image and original image to the regulation of each code coefficient, to overcome error.
WithThe autocorrelation matrix of shared dictionary is represented, is usedRepresent the autocorrelation matrix of each proprietary dictionary.WithRepresent the dictionary D={ D0,D1,D2,D3,D4,D5,D6Remaining each word after Zhong-1 block any one sub- dictionary
The connection matrix of allusion quotation.
Autocorrelation matrix adjustment factor sequence is represented with η, including:
Shared dictionary D0Autocorrelation matrixAdjustment factor η0, 0<η0< < 1, each proprietary sub- dictionary DkFrom phase
Close matrixAdjustment factor ηk, k=1,2 ..., 6,0<ηk< < 1.
With the cross correlation matrix number between each sub- dictionary in η ' expression dictionaries DAdjustment factor, k '=0,
1,2,…,6,0<ηk′' < < 1,
Being represented with λ carries out the regularization coefficient of regularization, 0 to the dictionary encoding coefficient matrices A<λ < < 1.
Step (2), from the XiIt is middle to extract the K0And K:
Step (2.1), randomly selects the equal training image of arbitrary size from image data base, each Zhang Suoshu instructions
It is setting quantity and equal-sized image fritter x to practice image cuttingi, x is expressed as with vector formi∈Rm;
Step (2.2), asks for each described xiIn each pixel j horizontal and vertical directions gradient gj, obtain described
xiGradient matrix Gi, abbreviation Gi, then by formula Gi=U Σ WTSingular value decomposition is carried out, is obtained:
M × m rank unitary matrice U, positive semidefinite m × 2 rank diagonal matrix Σ ∈ Rm×2, the element on diagonal is referred to as singular value, its
In nonzero element represent the energy size of corresponding gradient direction;2 × 2 rank unitary matrice W, W ∈ R2×2, each column vector represents institute
State xiThe gradient direction of interior each pixel;
Step (2.3), the difference parameter of gradient direction energy is represented with ρ, target xiMiddle all directions energy relative equilibrium
And relatively more stable subimage block is used as isotropic image block K0, as the publicly-owned feature;And it is obvious having
The subimage block of directionality as the anisotropic image block K, as proprietary feature, according to gradient direction angle ω, the K
It is divided into the gradient direction angular zone described in 6,ρ value, Σ between (0,1)1,1,Σ2,2It is two ladders
The degree each self-corresponding singular value in direction;
Step (3), to the shared dictionary D0With proprietary dictionary DkCombined optimization training is carried out according to the following steps:
If the object function of combined optimization is:
Wherein, djIt is unit vector, representing matrix DiColumn vector.Expression is asked so that object function is minimum
Variables Di,AiValue;[D0,Dk] represent matrix D0With DkIt is horizontally-spliced into a big matrix;||.||FRepresent square
The Frobenius norms of battle array;||.||0The L of representing matrix0The number of norm, i.e. nonzero element;Qk′Correspond to auto-correlation square
Battle arrayUnit matrix;||.||2It is the Euclid norm of vector.
Step (3.1), with recurrence least square dictionary learning algorithm RLA_DLA to the D0And DkInitialized.It is i.e. right
The image block K of the isotropic0Dictionary learning is carried out, the shared dictionary D for being initialized0, and its corresponding initialization
Code coefficient matrix A0To the image block K of the anisotropick, k=1,2 ..., 6, dictionary learning is carried out respectively, initialized
Proprietary sub- dictionary Dk, k=1,2 ..., 6, and its corresponding initialization code coefficient matrix
Step (3.2), solves the proprietary dictionary DkAnd it is corresponding
Step (3.2.1), changes the expression-form of the object function of the combined optimization training occurred in step (3):Find out
And ignore constant term, retain Dk,, alternately solve.
Step (3.2.1.1),Represent the isotropic image collection K0With shared dictionary D0Enter
Residual error after row sparse representation, this is constant, and object function value is not influenceed, negligible;
Step (3.2.1.2), due to A={ A0,AK},Thus in λ | | A | |0Xiang Zhong, ignores A0And, just it is reduced to
Step (3.2.1.3), rewrites
It is written over item and represents KkAnisotropy subimage block K in individual gradient direction angular zonekWith corresponding code coefficient
MatrixShared dictionary D after regulation0With corresponding code coefficient matrixProprietary sub- dictionary D after regulationkCarry out jointly
Residual error after sparse representation.Similarly, can handleItem is rewritten asOrderBeing written over the expression formula of result because obtained from is:
Step (3.2.1.3), the object function of the combined optimization training after being changed:
Wherein, k=1,2 ..., 6, λ, ηk,η′kIt is setting value.
Step (3.2.2), solves the code coefficient matrix of each proprietary sub- dictionary
Step (3.2.2.1), only retains and the code coefficientRelated item, ignores other,
Step (3.2.2.2), in DkUnder conditions of constant, obtain for solving the code coefficient matrixTarget letter
Number:
Step (3.2.2.3), sets degree of rarefication L, according to known KkAnd D0, tried to achieve with orthogonal matching pursuit algorithm OMPSo as to obtain Yk.Obtain corresponding in the object function for bringing step (3.2.2.2) into
Step (3.2.3), solves proprietary dictionary DK=[D1,D2,…,D6]:
Step (3.2.3.1), in the object function of step (3.2.1.3) the combined optimization training, ignores constant termRetain and each proprietary sub- dictionary DkRelated item, obtains:
Step (3.2.3.2), clicks each proprietary sub- dictionary D ' after step obtains combined optimization trainingk:
Step (3.2.3.3) makes:dγRepresent DkColumn vector, referred to as dictionary atom, γ is the sequence number of the column vector,
Step (3.2.3.4), d is updated with following gradient descent algorithmsγ, the d that γ is obtained after updatingγ' sub- the dictionary for constituting
With D 'kRepresent:
K=1,2 ..., 6, symbolRepresent to variable derivation, aγRepresent coefficient matrixγ rows, obtain:ζ1It is step-length.Step-length ζ1Determined by armijo criterions, the armi jo criterions, be linear search step
Algorithm long.
Step (3.3), solves shared dictionary D0:
Step (3.3.1), changes the expression-form of the object function of combined optimization training in step (3):
Step (3.3.1.1), each DkAnd the auto-correlation adjustment factor sequence η of correlationk, cross-correlation adjustment factor η 'k, respectively
Proprietary sub- dictionary encoding coefficient matrixIt is considered as constant, and A={ A0,AK},Retain D0,
Step (3.3.1.2) rewrites following every expression formulas
It is rewritten as:
It is rewritten as:
It is rewritten asWith
λ||A||0It is rewritten asAlso one λ | | A0||0。
Step (3.3.1.3), the revised combined optimization training objective function is:
Step (3.3.1.4), A is solved by step (3.3.1.3)0:
Specialized dictionary D ' after fixed renewalk, make Dk=D 'k, keepD0It is constant, obtain for solving A0It is described
Combined optimization training objective function:
A is tried to achieve with the orthogonal matching pursuit algorithm OMP algorithms described in step (3.2.2.3)0。
Step (3.3.1.5), the combined optimization training objective function proposed according to step (3.3.1.3) is solved
In Zk,A0,D0Under conditions of constant, exceptWithIt is able to retain outer, ignores other
, obtain for solvingThe combined optimization training objective function:
Obtained with step (3.3.1.4) identical method
Step (3.3.1.6), the combined optimization training objective function proposed by step (3.3.1.3) solves D0:
Step (3.3.1.6.1), the specialized dictionary D ' after fixed renewalk, retain and shared dictionary D0Related item, obtains
To solution D0The combined optimization training objective function:
Wherein, D-0The horizontally-spliced matrix of all of proprietary sub- dictionary is represented, its expression formula is D-0=[D '1,D
′2,…,D′6]。
Step (3.3.1.6.2), it is described updated proprietary with the gradient descent algorithm described in step (3.2.3.4)
Sub- dictionary Dk' and atom dγ', the atom after updating again is designated as d "γ, obtain:
Wherein:RepresentIn γ ' OK, ζ2Represent step-length, the armijo criterions described in step (3.2.3.4)
It is determined that.
Brief description of the drawings
Fig. 1, many geometry dictionary method for compressing image system block diagrams.
Fig. 2, combined optimization training program FB(flow block).
Fig. 3, image block energy differences parameter ρ probability distribution statistical figure, abscissa represents energy differences, and ordinate represents figure
As number of blocks.
Fig. 4, trains the dictionary for completing:
Fig. 4 .1, share dictionary, Fig. 4 .2, the specialized dictionary in part.
Fig. 5, Lena compression of images objective evaluation index:
Fig. 5 .1, coefficient nonzero value number,
Fig. 5 .2, rate-distortion curve.
Fig. 6, the image that image woman is encoded in code check for 0.15bpp:
Fig. 6 .1, JPEG (PSNR=28.68dB, SSIM=0.63),
Fig. 6 .2, JPEG2000 (PSNR=29.29dB, SSIM=0.77),
Fig. 6 .3, RLS_DLA (PSNR=28.5dB, SSIM=0.73),
Fig. 6 .4, the present invention, PSNR=29.76dB, SSIM=0.77).
Specific embodiment
Berkeley Segmentation Image database are chosen as training image collection, 200 are randomly selected
In image 8 × 104Used as training set, the size of each image block is 8 × 8 to individual image block.Test image comes from USC-SIPI
Data set, including some standard pictures, such as Lena, boat, man, couple, camera man, woman etc..What training was obtained
The size of dictionary is 200 dimensions, i.e., each dictionary has 200 atoms.The distribution situation of the energy differences parameter of image block.Will test
Image block seeks gradient, counts the energy difference of the main gradient direction of the Energy distribution of the gradient direction of each image block, i.e., two
Value, its probability-distribution function is as shown in Figure of description 2.The specialized dictionary of shared dictionary and part that training is obtained is illustrated in
In bright book accompanying drawing 3.
Experiment parameter is as shown in Table 1.
The experiment parameter of table 1.
Image compression standard JPEG, JPEG2000 methods, and dictionary learning algorithm RLS_DLA are chosen, K-SVD is used as right
Ratio method is tested.
By taking Lena images as an example, the number of the nonzero value corresponding to its dictionary, and rate-distortion curve is displayed in specification
In accompanying drawing 4.By taking woman images as an example, it is encoded when bit rate is 0.15bpp, and it is attached that reconstruction image is displayed in specification
In Fig. 5.Other test images are encoded in 0.5bpp and 0.4bpp, and its objective evaluating index PSNR is displayed in table 2.
The encoding efficiency when code check of table 2. is 0.5bpp (up) and 0.4bpp (descending)
It can be inferred that institute's pressure-raising compression method can obtain the more high-quality image of relative other method in low bit- rate,
Details retains more intact, and distortion rate is relatively low.
Claims (1)
1. image sparse characterizes the combined optimization training method of many dictionary learnings, it is characterised in that be one kind by learning image
In isotropic images the general character shared dictionary for obtaining and the proprietary dictionary that is obtained by learning the individual character in image carry out
Combined optimization training method, to realize that compression image while ensureing containing abundant information amount, can be improved to greatest extent again
Compression ratio, realizes according to the following steps successively in a computer:
Step (1), computer initialization sets following each parameters and coefficient:
Each image fritter will be obtained after training image X cuttingsI is xiSum, xi∈
Rm, m is the row of matrix R, abbreviation xi,
Image fritter xiIn pixel useRepresent, write a Chinese character in simplified form into j, j=1,2 ..., J, meanwhile, j is also the sequence number of pixel, and J is picture
The sum of element, the horizontal component of the pixel is represented with u, and vertical component is represented with v;
Use gjThe gradient of the pixel j is represented,Use GiRepresent described image fritter xiGradient
Matrix, Gi=[g1,g2,…,gJ]T,Gi∈RJ×2, columns n=2 represent the number J of pixel gradient, row expression pixel gradient with row
Level and vertical component, use respectively Represent;
Deflections of the pixel j in gradient direction is represented with ω
Use K0Represent that the collection of isotropism subimage block is combined into, abbreviation K0, use D0Represent study K0The shared dictionary for obtaining;
The set of anisotropic image block, K={ K are represented with K1,K2,…,Kk,…,K6}={ Kk, referred to as 6 class gradient directions
The region k at angle, wherein, K1Correspondence (0 °, 30 °), K2Correspondence (30 °+, 60 °), K3Correspondence (60 °+, 90 °), K4Correspondence (90 °+,
120 °), K5Correspondence (120 °+, 150 °), K6Correspondence (150 °+, 180 °), wherein, 30 °+represent that the rest may be inferred by analogy for it, symbol more than 30 °
" { } " expression " contains " element, similarly hereinafter;
Use DKRepresent the proprietary dictionary obtained after study K, abbreviation DK, correspondingly, use D1~D66 institutes of corresponding study are represented respectively
The 6 proprietary sub- dictionaries obtained after the anisotropy image block for stating gradient angular zone, and DKIt is made up of 6 proprietary sub- dictionaries, DK=
{D1,D2,…,D6}={ Dk};
The shared dictionary D is represented with D0With proprietary sub- dictionary DkSet, be expressed as:D={ D0, DK}={ D0,D1,D2,D3,
D4,D5,D6};
With k=1,2 ..., 6 represent the subscript sequence of each proprietary sub- dictionaries;Institute in dictionary D is represented with k '=0,1,2,3,4,5,6
The subscript sequence of the sub- dictionary of whole for containing;
Dictionary encoding coefficient matrix, A={ A are represented with A0,AK, wherein:A0Represent the shared dictionary D0Code coefficient matrix,
AKRepresent the proprietary dictionary DKCode coefficient matrix, wherein,Represent AKIncluding:Proprietary sub- wordbook
Close { DkIn correspond to the set of anisotropy subimage block in each general character image part dictionary code coefficientAnd
Each proprietary sub- dictionary D of 6 class anisotropy subgraphskCode coefficient matrix
The element of each code coefficient matrix is referred to as code coefficient a, δ<A < < 1, set code coefficient as combined optimization is trained
One of parameter, its object is to:In follow-up study, when judge first through step (2.3) in target obtain, afterwards again through parameter
When the compression image of variable combined optimization is unsatisfactory for error requirements with the original input image of reality on image similarity, can pass through
The similarity of compression image and original image is improved to the regulation of each code coefficient, to overcome error;
WithThe autocorrelation matrix of shared dictionary is represented, is usedRepresent the autocorrelation matrix of each proprietary dictionary;WithRepresent the dictionary D={ D0,D1,D2,D3,D4,D5,D6Remaining each word after Zhong-1 block any one sub- dictionary
The connection matrix of allusion quotation;
Autocorrelation matrix adjustment factor sequence is represented with η, including:
Shared dictionary D0Autocorrelation matrixAdjustment factor η0, 0<η0< < 1, each proprietary sub- dictionary DkAuto-correlation square
Battle arrayAdjustment factor ηk, k=1,2 ..., 6,0<ηk< < 1;
With the cross correlation matrix number between each sub- dictionary in η ' expression dictionaries DAdjustment factor, k '=0,1,
2,…,6,0<ηk′' < < 1,
Being represented with λ carries out the regularization coefficient of regularization, 0 to the dictionary encoding coefficient matrices A<λ < < 1;
Step (2), from the XiIt is middle to extract the K0And K:
Step (2.1), randomly selects the equal training image of arbitrary size from image data base, each Zhang Suoshu is trained and is schemed
As cutting is setting quantity and equal-sized image fritter xi, x is expressed as with vector formi∈Rm;
Step (2.2), asks for each described xiIn each pixel j horizontal and vertical directions gradient gj, obtain the xi's
Gradient matrix Gi, abbreviation Gi, then by formula Gi=U Σ WTSingular value decomposition is carried out, is obtained:
M × m rank unitary matrice U, positive semidefinite m × 2 rank diagonal matrix Σ ∈ Rm×2, the element on diagonal is referred to as singular value, therein
Nonzero element represents the energy size of corresponding gradient direction;2 × 2 rank unitary matrice W, W ∈ R2×2, each column vector represents the xi
The gradient direction of interior each pixel;
Step (2.3), the difference parameter of gradient direction energy is represented with ρ, target xiMiddle all directions energy relative equilibrium and compare
Stable subimage block is used as isotropic image block K0, as the publicly-owned feature;And with obvious directionality
Subimage block as the anisotropic image block K, be 6 the K points according to gradient direction angle ω as proprietary feature
Individual described gradient direction angular zone,ρ value, Σ between (0,1)1,1,Σ2,2It is two gradient directions
Each self-corresponding singular value;
Step (3), to the shared dictionary D0With proprietary dictionary DkCombined optimization training is carried out according to the following steps:
If the object function of combined optimization is:
Wherein, djIt is unit vector, representing matrix DiColumn vector;The change for causing that object function is minimum is asked in expression
Amount Di,AiValue;[D0,Dk] represent matrix D0With DkIt is horizontally-spliced into a big matrix;||.||FRepresenting matrix
Frobenius norms;||.||0The L of representing matrix0The number of norm, i.e. nonzero element;Qk′Correspond to autocorrelation matrixUnit matrix;||.||2It is the Euclid norm of vector;
Step (3.1), with recurrence least square dictionary learning algorithm RLA_DLA to the D0And DkInitialized;I.e. to described
The image block K of isotropic0Dictionary learning is carried out, the shared dictionary D for being initialized0, and its corresponding initialization coding
Coefficient matrices A0To the image block K of the anisotropick, k=1,2 ..., 6, dictionary learning is carried out respectively, what is initialized is special
There is sub- dictionary Dk, k=1,2 ..., 6, and its corresponding initialization code coefficient matrix
Step (3.2), solves the proprietary dictionary DkAnd it is corresponding
Step (3.2.1), changes the expression-form of the object function of the combined optimization training occurred in step (3):Find out and neglect
Slightly constant term, retains Dk,, alternately solve;
Step (3.2.1.1),Represent the isotropic image collection K0With shared dictionary D0Carry out sparse
Residual error after sign, this is constant, and object function value is not influenceed, negligible;
Step (3.2.1.2), due toThus in λ | | A | |0Xiang Zhong, ignores A0And, just it is reduced to
Step (3.2.1.3), rewrites
It is written over item and represents KkAnisotropy subimage block K in individual gradient direction angular zonekWith corresponding code coefficient matrixShared dictionary D after regulation0With corresponding code coefficient matrixProprietary sub- dictionary D after regulationkCarry out jointly sparse
Residual error after sign;Similarly, can handleItem is rewritten asOrderBeing written over the expression formula of result because obtained from is:
Step (3.2.1.3), the object function of the combined optimization training after being changed:
Wherein, k=1,2 ..., 6, λ, ηk, η 'kIt is setting value;
Step (3.2.2), solves the code coefficient matrix of each proprietary sub- dictionary
Step (3.2.2.1), only retains and the code coefficientRelated item, ignores other,
Step (3.2.2.2), in DkUnder conditions of constant, obtain for solving the code coefficient matrixObject function:
Step (3.2.2.3), sets degree of rarefication L, according to known KkAnd D0,
Tried to achieve with orthogonal matching pursuit algorithm OMPSo as to obtain Yk;Obtained in the object function for bringing step (3.2.2.2) into
It is corresponding
Step (3.2.3), solves proprietary dictionary DK=[D1,D2,…,D6]:
Step (3.2.3.1), in the object function of step (3.2.1.3) the combined optimization training, ignores constant termRetain and each proprietary sub- dictionary DkRelated item, obtains:
Step (3.2.3.2), clicks each proprietary sub- dictionary D ' after step obtains combined optimization trainingk:
Step (3.2.3.3) makes:dγRepresent DkColumn vector, referred to as dictionary atom, γ is the sequence number of the column vector,
Step (3.2.3.4), d is updated with following gradient descent algorithmsγ, the d that γ is obtained after updatingγ' sub- dictionary the D ' for constitutingk
Represent:
K=1,2 ..., 6, symbolRepresent to variable derivation, aγRepresent coefficient matrixγ rows, obtain:ζ1It is step-length;Step-length ζ1Determined by armijo criterions, the armi jo criterions, be linear search step
Algorithm long;
Step (3.3), solves shared dictionary D0:
Step (3.3.1), changes the expression-form of the object function of combined optimization training in step (3):
Step (3.3.1.1), each DkAnd the auto-correlation adjustment factor sequence η of correlationk, cross-correlation adjustment factor η 'k, it is each proprietary
Sub- dictionary encoding coefficient matrixIt is considered as constant, andRetain
Step (3.3.1.2) rewrites following every expression formulas
It is rewritten as:
It is rewritten as:
It is rewritten asWith
λ||A||0It is rewritten asAlso one λ | | A0||0;
Step (3.3.1.3), the revised combined optimization training objective function is:
Step (3.3.1.4), A is solved by step (3.3.1.3)0:
Specialized dictionary D ' after fixed renewalk, make Dk=D 'k, keepD0It is constant, obtain for solving A0The joint
Optimization training objective function:
A is tried to achieve with the orthogonal matching pursuit algorithm OMP algorithms described in step (3.2.2.3)0;
Step (3.3.1.5), the combined optimization training objective function proposed according to step (3.3.1.3) is solved
In Zk,A0,D0Under conditions of constant, exceptWithIt is able to retain outer, ignores other,
Obtain for solvingThe combined optimization training objective function:
Obtained with step (3.3.1.4) identical method
Step (3.3.1.6), the combined optimization training objective function proposed by step (3.3.1.3) solves D0:
Step (3.3.1.6.1), the specialized dictionary D ' after fixed renewalk, retain and shared dictionary D0Related item, is asked
Solution D0The combined optimization training objective function:
Wherein, D-0The horizontally-spliced matrix of all of proprietary sub- dictionary is represented, its expression formula is D-0=[D '1,D′2,…,
D′6];
Step (3.3.1.6.2), with the gradient descent algorithm described in step (3.2.3.4), the updated proprietary sub- word
Allusion quotation Dk' and atom dγ', the atom after updating again is designated as d "γ, obtain:
Wherein:RepresentIn γ ' OK, ζ2Step-length is represented, armi jo criterions are true described in step (3.2.3.4)
It is fixed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611252617.8A CN106815876B (en) | 2016-12-30 | 2016-12-30 | Image sparse characterizes the combined optimization training method of more dictionary learnings |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611252617.8A CN106815876B (en) | 2016-12-30 | 2016-12-30 | Image sparse characterizes the combined optimization training method of more dictionary learnings |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106815876A true CN106815876A (en) | 2017-06-09 |
CN106815876B CN106815876B (en) | 2019-08-02 |
Family
ID=59109243
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611252617.8A Active CN106815876B (en) | 2016-12-30 | 2016-12-30 | Image sparse characterizes the combined optimization training method of more dictionary learnings |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106815876B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110678899A (en) * | 2017-06-02 | 2020-01-10 | 三菱电机株式会社 | Imaging system and image processing method |
CN111666967A (en) * | 2020-04-21 | 2020-09-15 | 浙江工业大学 | Image classification method based on incoherent joint dictionary learning |
CN112163615A (en) * | 2020-09-25 | 2021-01-01 | 电子科技大学 | Unknown target discrimination method based on threshold optimization dictionary learning |
US11122095B2 (en) | 2019-09-23 | 2021-09-14 | Netapp, Inc. | Methods for dictionary-based compression and devices thereof |
CN114063156A (en) * | 2020-07-30 | 2022-02-18 | 中国石油天然气股份有限公司 | Seismic anisotropy gradient inversion method and system based on sparse representation |
CN116227367A (en) * | 2023-05-10 | 2023-06-06 | 广东电网有限责任公司珠海供电局 | Back pressure prediction model construction method, back pressure prediction method and back pressure prediction device of direct air cooling system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103489203A (en) * | 2013-01-31 | 2014-01-01 | 清华大学 | Image coding method and system based on dictionary learning |
CN104159112A (en) * | 2014-08-08 | 2014-11-19 | 哈尔滨工业大学深圳研究生院 | Compressed sensing video transmission method and system based on dual sparse model decoding |
CN104392251A (en) * | 2014-11-28 | 2015-03-04 | 西安电子科技大学 | Hyperspectral image classification method based on semi-supervised dictionary learning |
US9025883B2 (en) * | 2013-07-01 | 2015-05-05 | Xerox Corporation | Adaptive quality image reconstruction via a compressed sensing framework |
CN106658003A (en) * | 2016-09-27 | 2017-05-10 | 清华大学 | quantization method of dictionary learning-based image compression system |
-
2016
- 2016-12-30 CN CN201611252617.8A patent/CN106815876B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103489203A (en) * | 2013-01-31 | 2014-01-01 | 清华大学 | Image coding method and system based on dictionary learning |
US9025883B2 (en) * | 2013-07-01 | 2015-05-05 | Xerox Corporation | Adaptive quality image reconstruction via a compressed sensing framework |
CN104159112A (en) * | 2014-08-08 | 2014-11-19 | 哈尔滨工业大学深圳研究生院 | Compressed sensing video transmission method and system based on dual sparse model decoding |
CN104392251A (en) * | 2014-11-28 | 2015-03-04 | 西安电子科技大学 | Hyperspectral image classification method based on semi-supervised dictionary learning |
CN106658003A (en) * | 2016-09-27 | 2017-05-10 | 清华大学 | quantization method of dictionary learning-based image compression system |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110678899A (en) * | 2017-06-02 | 2020-01-10 | 三菱电机株式会社 | Imaging system and image processing method |
CN110678899B (en) * | 2017-06-02 | 2023-04-28 | 三菱电机株式会社 | Imaging system and image processing method |
US11122095B2 (en) | 2019-09-23 | 2021-09-14 | Netapp, Inc. | Methods for dictionary-based compression and devices thereof |
US11444997B2 (en) | 2019-09-23 | 2022-09-13 | Netapp, Inc. | Methods for dictionary-based compression and devices thereof |
CN111666967A (en) * | 2020-04-21 | 2020-09-15 | 浙江工业大学 | Image classification method based on incoherent joint dictionary learning |
CN111666967B (en) * | 2020-04-21 | 2023-06-13 | 浙江工业大学 | Image classification method based on incoherence combined dictionary learning |
CN114063156A (en) * | 2020-07-30 | 2022-02-18 | 中国石油天然气股份有限公司 | Seismic anisotropy gradient inversion method and system based on sparse representation |
CN112163615A (en) * | 2020-09-25 | 2021-01-01 | 电子科技大学 | Unknown target discrimination method based on threshold optimization dictionary learning |
CN112163615B (en) * | 2020-09-25 | 2022-04-22 | 电子科技大学 | Unknown target discrimination method based on threshold optimization dictionary learning |
CN116227367A (en) * | 2023-05-10 | 2023-06-06 | 广东电网有限责任公司珠海供电局 | Back pressure prediction model construction method, back pressure prediction method and back pressure prediction device of direct air cooling system |
Also Published As
Publication number | Publication date |
---|---|
CN106815876B (en) | 2019-08-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106815876B (en) | Image sparse characterizes the combined optimization training method of more dictionary learnings | |
CN111787323B (en) | Variable bit rate generation type compression method based on counterstudy | |
CN104159112B (en) | The compressed sensing video transmission method and system decoded based on dual sparse model | |
CN106911930A (en) | It is a kind of that the method for perceiving video reconstruction is compressed based on recursive convolution neutral net | |
CN107689034A (en) | A kind of training method of neutral net, denoising method and device | |
CN102156875A (en) | Image super-resolution reconstruction method based on multitask KSVD (K singular value decomposition) dictionary learning | |
CN104867119B (en) | The structural missing image fill method rebuild based on low-rank matrix | |
CN107067367A (en) | A kind of Image Super-resolution Reconstruction processing method | |
CN108121975A (en) | A kind of face identification method combined initial data and generate data | |
CN107154064B (en) | Natural image compressed sensing method for reconstructing based on depth sparse coding | |
CN109523513A (en) | Based on the sparse stereo image quality evaluation method for rebuilding color fusion image | |
CN103338379B (en) | Stereoscopic video objective quality evaluation method based on machine learning | |
CN111047661B (en) | CS-MRI image reconstruction method based on sparse manifold joint constraint | |
CN106952317A (en) | Based on the high spectrum image method for reconstructing that structure is sparse | |
CN104376565A (en) | Non-reference image quality evaluation method based on discrete cosine transform and sparse representation | |
CN106529586A (en) | Image classification method based on supplemented text characteristic | |
CN117597703A (en) | Multi-scale converter for image analysis | |
CN105160653A (en) | Quality evaluation method for fog-degraded images | |
CN103473744A (en) | Variable weight type compressed sensing sampling based image reconstruction method allowing space domain to be reduced | |
CN109146785A (en) | A kind of image super-resolution method based on the sparse autocoder of improvement | |
CN104408697A (en) | Image super-resolution reconstruction method based on genetic algorithm and regular prior model | |
CN108108652A (en) | A kind of across visual angle Human bodys' response method and device based on dictionary learning | |
CN107146259A (en) | A kind of New Image compression method based on compressive sensing theory | |
CN105118025B (en) | Rapid image ultra-resolution method based on soft-threshold coding | |
CN116403239A (en) | Human body posture estimation method based on transducer space-time characteristic enhancement |
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 |