CN103034970A - Multiple information hiding method based on combination of image normalization and principal component analysis (PCA) - Google Patents

Abstract

The invention discloses an information hiding method based on the combination of image normalization and principal component analysis (PCA). The information hiding method belongs to the technical area of a computer image processing and information security. Based on the theory of image normalization and invariant centroid, the information hiding method has a good resistance to the geometric attack. Before implanting the hidden information, the chaotic technology is utilized to carry out the encryption processing on the hidden images so that the confidentiality and security of the hidden information are effectively improved. According to the features of the visual system---illuminance masking and texture masking, a perceptual masking template is designed to determine the intensity factor embedded by each sub-block hidden information and then the hidden information is embedded into multiple subspaces of the images so that the information embedding capacity of the cover images can be effectively improved.

Description

A kind of many information concealing methods that combine based on image normalization and PCA

Technical field

The present invention relates to a kind of many information concealing methods that combine based on image normalization and PCA, belong to Computer Image Processing and field of information security technology.

Background technology

The ultimate principle of image information concealing technology is to utilize in the image information ubiquitous redundancy to wherein embedding secret information, thereby reaches the purpose of hidden important information.It is mainly studied and how to be embedded into secret data in the view data and the problem of How to choose embedded location, key issue is that information embeds the design of algorithm, and how to process the robustness that hides Info, the relation between sentience and embedded this three of data volume not.

Information Hiding Algorithms early all is on the spatial domain, this class algorithm is by directly revising carrier numerical information, pixel such as image, thereby directly be carried in secret information on the data, its advantage is quick, and possesses certain resistivity for operations such as the geometric transformation of carrier image, compressions.Although the method ratio that carries out Information hiding by the spatial domain algorithm is easier to realize, the minimum modification of concealed carrier is all had very large fragility, if the assailant wants to destroy secret information, only need use simply signal processing technology and just can accomplish.In many cases, even lossy compression method also can cause losing of information, and has more robustness in signal frequency domain embedding information than in time domain embedding information, and it all is that running is on certain frequency domain that existing robustness is hidden system's reality preferably.Transform-domain algorithm has the ability that better opposing compression, cutting and some other image processing method are attacked than spatial domain algorithm, has also kept simultaneously the disguise to human vision.

The proper subspace mapping algorithm is a kind of information concealing method that development in recent years is got up, the method is by decomposing digital Characteristic of Image space, in each sub spaces that is embedded into image with hiding Info, can effectively realize hiding of many information, and have preferably robustness.

Proper subspace can be divided into signal subspace and noise subspace, because human eye is not too responsive to little noise, Information hiding can be strengthened its not sentience in noise subspace.Simultaneously, because the mutual independence between the different subspace carries out the data volume that Information hiding can increase embedding in different subspace, and can realize hiding of many information based on different subspace.In recent years, also be subject to people based on the information concealing method of proper subspace and more and more pay attention to, and become gradually a kind of important information concealing method.

Principal component analysis is a kind of very important proper subspace extracting method, its objective is the unit orthogonal vector base (being pivot) of seeking one group of optimum by linear transformation, and come reconstruction sample with the linear combination of part vector wherein, make sample and the error of former sample under the lowest mean square meaning after the reconstruction minimum.Because the orthogonality between the subspace, information in the different subspace will independently exist, do not interfere with each other, therefore, under the prerequisite of not damaging picture quality, can come embedding information to improve the information embedding capacity of carrier image by the number that increases the subspace, the method is processed to attack to some conventional signals has good robustness.Yet, the same with most image digitization information concealing method, Information Hiding in Digital Image based on principal component analysis also lacks enough resistibilitys to geometric attack, when the geometric transformation of the routines such as image translation, rotation or convergent-divergent occurring, just be difficult to hide Info from carrier image, to extract.

Image normalization is exactly the normalized parameter by computed image, then passes through a series of conversion, convert pending original image to the canonical form image, and this canonical form image has invariant feature to translation, rotation, upset, the conversion of convergent-divergent equiaffine.At present, the geometric moment that most image normalization parameter all is based on image calculates, the transformation parameter of determining by geometric invariant moment can be normalized to the canonical form image to the original image through certain uncertain affined transformation, and the method has a wide range of applications in the fields such as computer vision, pattern-recognition.And in the digital watermarking field, the image normalization technology mainly is applied to watermark synchronization, namely first image is carried out normalization before watermark embedding and watermark extracting, and with the impact of elimination geometric transformation, and then effective opposing is to the geometric attack of digital watermarking.

Summary of the invention

The present invention is directed to the proposition of above problem, and develop a kind of many information concealing methods that combine based on image normalization and PCA, it has effectively merged the resist geometric attacks ability of image normalization and PCA and can increase arbitrarily the subspace number and increase to hide Info and do not affect the data compression of robustness ability, based on image normalization and the psychological opinion of never degenerating, a kind of many information concealing methods that combine based on image normalization and PCA have been proposed.

The present invention takes following steps:

Step 1: be 256 grades of gray level image F={f (i, j) for initial carrier, 1≤i≤m, 1≤j≤m}, the image size is m * m, hiding Info is bianry image W={w (i, j), 1≤i≤s, 1≤j≤s}, the image size is s * s; Carrier image is carried out normalized based on geometric invariant moment, try to achieve normalized image I and image centroid thereof, and corresponding transformation parameter R;

Step 2, get and embed the zone and with its piecemeal;

Step 3, determine the intensity factor that hides Info;

Step 4, take the initial parameter that generates the Logistic chaos sequence as encryption key, carry out Chaotic Scrambling to hiding Info;

Step 5, calculating S _tThe KL conversion coefficient; Step 6, use amended sub-space feature vectors

Replace original u _i, carry out the sample vector after reconstruct is tried to achieve in the KL inverse transformation, add the average μ that deducts in the step 2 ₀, and be transformed to the image subblock of s * s, then be rearranged into and embed the area image that hides Info;

Step 7, hide Info area image and original image remainder of embedding is merged into and contains the carrier image that hides Info, carry out contrary normalization to containing the carrier image that hides Info by transformation parameter R, obtain to embed the carrier image F' after hiding Info, finish the embedding that hides Info;

The extraction of step 8, image information;

Key K (μ, x in step 9, the employing hidden image scramble process ₀), regenerate chaos sequence, then with the inverse process of scramble process, to the scramble that the extracts random conversion that is inverted that hides Info, recover the EW that hides Info.

On the basis of traditional PCA Information Hiding Techniques, be the geometric attack of opposing image, the initial carrier image has been adopted image normalization technology based on geometric invariant moment, realize the geometry correction of image.

Step 3 is described to be determined to it is characterized in that the intensity factor that hides Info and embed: adopting the PCA algorithm to ask between the Image Subspace, adopt and shelter and texture masking based on the illumination of vision system, design perceptual mask template: α=c ₁(1-NVF)+c ₂NVF determines the intensity factor that each sub-block hides Info and embeds.

The principle of the invention: the present invention proposes a kind of many information concealing methods that combine based on image normalization and PCA, the method is before the information of carrying out embeds, at first carry out chaos encryption processes hiding Info, then adopt the impact of eliminating translation, rotation, upset, the conversion of convergent-divergent equiaffine based on the image normalization of geometric invariant moment, make it have affine unchangeability.Then, the normalized image barycenter is carried out piecemeal on every side, and adopt the PCA method to ask for its sample vector and total population scatter matrix.Afterwards, in order to obtain preferably image perception quality and higher robustness, this method is sheltered and texture masking based on the illumination of vision system, determines that according to the local characteristics of different sub-blocks each sub-block embeds the intensity factor that hides Info.The invisibility that hides Info in order to regulate adaptively the intensity and guaranteeing of hiding Info utilizes human visual system's (Humanvisual system, HVS) characteristic to design the perceptual mask template.At last, embed the intensity factor that hides Info based on each sub-block, the hidden image behind the scramble is embedded in the subspace of block image, realize Information hiding.Equally, before information extraction, the image that is subject to hiding Info containing of geometric attack is carried out inverse transformation based on geometric invariant moment, extract watermark with the method for principal component analysis again

The present invention compared with prior art has the following advantages:

1, because PCA utilizes the proper subspace mapping algorithm that the feature space of digital picture is decomposed, simultaneously because the mutual independence between the different subspace, in different subspace, carry out Information hiding and increased the data volume that embeds, thereby realized many Information hiding.Owing to be the image normalization that carried out based on geometric invariant moment, effectively eliminated the geometric transformation impact simultaneously before information embeds, well opposing is to the geometric attack of digital watermarking.In the leaching process that hides Info, the transformation matrix of use is identical with original image, and this not only can extract in identical subspace and hide Info, and has also increased the security that hides Info simultaneously.

2, in the past information concealing method all is on the spatial domain, mostly just information extraction from the original position that embeds, suffer the information extraction behind the geometric attack often can not be satisfactory, and all can produce very large fragility to the minimum modification of concealed carrier, in a lot of situations, even lossy compression method also can cause losing of information.Given this, we propose to carry out chaos encryption and process before information embeds, the image normalization of geometric invariant moment, eliminate the impact of geometric transformation, its key is will accurately calculate before the watermark extracting and is subject to the geometric attack parameter, and the watermarking images that contains that is subject to geometric attack is carried out inverse transformation, extracts watermark with the principal component analysis method again, the method is not only processed some conventional signals has good robustness, and geometric attack is also had good resistibility.In addition, because the good data compression that the PCA conversion has can effectively realize Data Dimensionality Reduction, reduce the complexity of calculating in the image processing.

Description of drawings

Fig. 1 is based on the embedding grammar process flow diagram that hides Info of principal component analysis.

Fig. 2 (a) original image hides Info.

Fig. 2 (b) initial carrier image.

Fig. 3 image normalization and embedding zone are selected.

Fig. 4 (a) initial carrier image.

Fig. 4 (b) contains the image that hides Info.

The enciphering hiding information that Fig. 4 (c) extracts.

What Fig. 4 (d) extracted hides Info.

Fig. 5 (a) initial carrier image

Fig. 5 (b) hides Info 1.

Fig. 5 (c) hides Info 2.

Fig. 5 (d) enciphering hiding information 1.

Fig. 5 (e) enciphering hiding information 2.

Fig. 5 (g) contains the image that hides Info.

Fig. 5 (i) extracts enciphering hiding information 1.

Fig. 5 (j) extracts enciphering hiding information 2.

Fig. 5 (k) extracts and hides Info 1.

Fig. 5 (l) extracts and hides Info 2.

Embodiment

Below in conjunction with accompanying drawing the present invention is done and further specifies, one embodiment of the present of invention are:

As shown in Figure 1: the process flow diagram of performing step of the present invention;

Step 1: be 256 grades of gray level image F={f (i for initial carrier, j), 1≤i≤m, 1≤j≤m}, the image size is m * m, shown in Fig. 2 (a), hiding Info is bianry image W={w (i, j), 1≤i≤s, 1≤j≤s}, the image size is s * s, shown in Fig. 2 (b); Carrier image is carried out normalized based on geometric invariant moment, try to achieve normalized image I and image centroid thereof, as shown in Figure 3, and corresponding transformation parameter R;

Described image normalization process such as following four steps based on geometric invariant moment:

(1) coordinate centralization

Original image f (x, y) is carried out the coordinate centralization according to following formula;

$\left(\begin{array}{c}{x}_a\\ y_a\end{array}\right)=\left(\begin{array}{cc}1& 0\\ 0& 1\end{array}\right)\left(\begin{array}{c}x\\ y\end{array}\right)-\left(\begin{array}{c}{d}_1\\ d_2\end{array}\right)---\left(1\right)$

Wherein,

m ₁₀, m ₀₁, m ₀₀Be the geometric moment of original image f (x, y), its computing formula is: $m_{\mathrm{pq}}=\underset{x=0}{\overset{N_1-1}{\mathrm{\&Sigma;}}}\underset{y=0}{\overset{N_2-1}{\mathrm{\&Sigma;}}}{x}^p{y}^{q}f(x,y),0\&le;x<m,0\&le;y<m,$ F (x _a, y _a) be the image after the coordinate centralization.

(2) x-shearing normalization

To the image f (x after the coordinate centralization _a, y _a) carry out conversion process according to following formula;

$\left(\begin{array}{c}{x}_b\\ y_b\end{array}\right)=\left(\begin{array}{cc}1& \mathrm{\&beta;}\\ 0& 1\end{array}\right)\left(\begin{array}{c}{x}_a\\ y_a\end{array}\right)---\left(2\right)$

Wherein, parameter μ ₁₁, μ ₀₂Centered by change image f (x _a, y _a) the center square, its computing formula is:

$u_{\mathrm{pq}}=\underset{x=0}{\overset{N_1-1}{\mathrm{\&Sigma;}}}\underset{y=0}{\overset{N_2-1}{\mathrm{\&Sigma;}}}{(x-\stackrel{\&OverBar;}{x})}^p{(y-\stackrel{\&OverBar;}{y})}^{q}f(x,y),0\&le;x<m,0\&le;y<m$

$\stackrel{\&OverBar;}{x}=\frac{m_{10}}{m_{00}},\stackrel{\&OverBar;}{y}=\frac{m_{01}}{m_{00}}---\left(3\right)$

F (x _b, y _b) image after the expression x-shearing normalization.

(3) convergent-divergent normalization

To the image f (x after the x-shearing normalization _by _b) carry out conversion process according to following formula;

$\left(\begin{array}{c}{x}_c\\ y_c\end{array}\right)=\left(\begin{array}{cc}\mathrm{\&alpha;}& 0\\ 0& \mathrm{\&delta;}\end{array}\right)\left(\begin{array}{c}{x}_b\\ y_b\end{array}\right)---\left(4\right)$

Described

μ ₂₀, μ ₀₂Be image f (x _b, y _b) the center square, f (x _c, y _c) image after the expression yardstick normalization;

(4) rotation normalization

To the image f (x after the convergent-divergent normalization _c, y _c) carry out conversion process according to following formula;

$\left(\begin{array}{c}{x}_d\\ y_d\end{array}\right)=\left(\begin{array}{cc}\cos \mathrm{\&phi;}& \sin \mathrm{\&phi;}\\ -\sin \mathrm{\&phi;}& \cos \mathrm{\&phi;}\end{array}\right)\left(\begin{array}{c}{x}_c\\ y_c\end{array}\right)---\left(5\right)$

Described

I=f (x _d, y _d) image after the expression rotation normalization.

Step 2, get and embed the zone and with its piecemeal;

Point centered by the barycenter of normalized image I at first is divided into piece image F N size and is the sub-block of s * s, each sub-block is rearranged the vector x that obtains n=s * s dimension _i, i=1,2 ..., N is zero in order to make its average, need to deduct its mean vector namely:

A＝(x ₁-μ ₀,x ₂-μ _0,...,x _N-μ ₀)（6）

A is called the training sample matrix in the formula,

Be the population mean of sample, at this moment, the autocorrelation matrix R among the KLT has become covariance matrix, and its maximal possibility estimation is total population scatter matrix:

$S_t=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}(x_i-{\mathrm{\&mu;}}_0){(x_i-{\mathrm{\&mu;}}_0)}^T---\left(7\right)$

Because AA ^TAnd A ^TA has identical eigenwert, and AA ^TCorresponding to eigenvalue λ _iProper vector u _iWith A ^TThe corresponding proper vector v of A _iHave following relationship:

$u_i=\frac{1}{\sqrt{{\mathrm{\&lambda;}}_i}}A{v}_i---\left(8\right)$

Then to the sample x among any X _iCan be expressed as:

$x_i=\underset{j=1}{\overset{n}{\mathrm{\&Sigma;}}}{y}_j{u}_j---\left(9\right)$

Y wherein _j=x _i ^Tu _j(j=1,2 ..., n), y=[y ₁, y ₂..., y _n] ^TChoose front d the component of y

As feature, then can demonstrate,prove Be d component of variance maximum (being that energy is maximum), and

At all with S _tThe reconstruct of d proper vector in have minimum square error, thereby

Be commonly called main composition, with

Corresponding subspace is called as signal subspace, and then obtains sample vector A and total population scatter matrix S _t

Step 3, determine the intensity factor that hides Info;

Design perceptual mask template is as follows:

α＝c ₁·(1-NVF)+c ₂·NVF（10）

Described c ₁And c ₂Be used for regulating the embedment strength that hides Info, c ₁Can be used to regulate the visual quality that contains the image that hides Info.Noise visible function (Noise visibility function, NVF) adopts following form:

$\mathrm{NVF}(x,y)=\frac{1}{1+k\&CenterDot;{\mathrm{\&sigma;}}_x^2(x,y)},k=\frac{D}{{\mathrm{\&sigma;}}_{x\max }^2}---\left(11\right)$

Wherein, k adjusts parameter, and every width of cloth image has different adjustment parameters,

The local variance of original image in the expression window;

The maximal value of expression original image local variance.

Step 4, take the initial parameter that generates the Logistic chaos sequence as encryption key, carry out Chaotic Scrambling to hiding Info, it comprises that step is as follows:

(1) input key K (μ, x ₀), produce chaos sequence X, X is arranged by ascending order, namely [X ', l]=sort (X), X' is the chaos sequence that ascending order is arranged, l is index sequence;

(2) W that will hide Info is converted to one-dimensional vector, then is rearranged to the scramble W' that hides Info with index sequence l.

Step 5, by formula (9) calculate S _tThe KL conversion coefficient, get its eigenwert y _i, i ∈ [1,2., N] corresponding sub-space feature vectors u _i, wherein, N is the sub-block number of cutting, then, and at u _iEmbed the scramble W' that hides Info, concrete formula is:

$u_{\mathrm{ij}}^w=u_{\mathrm{ij}}+{\mathrm{\&alpha;}}_j\&CenterDot;w_j^{\&prime;},j=\mathrm{1,2},...,s\&times;s---\left(12\right)$

Its intensity factor α _jAccording to the local characteristics of different sub-blocks by formula (10) calculate; Step 6, use amended sub-space feature vectors

Replace original u _i, carry out the sample vector after reconstruct is tried to achieve in the KL inverse transformation, add the average μ that deducts in the step (2) ₀, and be transformed to the image subblock of s * s, then be rearranged into and embed the area image that hides Info;

Step 7, hide Info area image and original image remainder of embedding is merged into and contains the carrier image that hides Info, carry out contrary normalization to containing the carrier image that hides Info by transformation parameter R, obtain to embed the carrier image F' after hiding Info, finish the embedding that hides Info;

The extraction of step 8, image information

Leaching process is as follows:

(1) utilization is treated detected image F' and is carried out normalized, to obtain corresponding normalized image I' based on the image normalization technology of square;

(2) with hide Info identically when embedding, equally the embedding of normalized image I' is hidden Info that extract in the zone and piecemeal, then be transformed to sample vector by formula (7) try to achieve its total population scatter matrix S _t';

KL transformation matrix when (3) utilization hides Info embedding is to S _t' carry out feature extraction, obtain corresponding to y _i, i ∈ [1,2 ..., N] sub-space feature vectors u _i';

Intensity factor when (4) embedding according to each image subblock, the setting threshold decision function embeds the extraction of the EW that hides Info, and formula is as follows:

$e{w}_i=\left\{\begin{array}{cc}1& u_i^{\&prime;}-u_i\&GreaterEqual;\mathrm{\&alpha;}\\ 0& u_i^{\&prime;}-u_i<\mathrm{\&alpha;}\end{array}\right.i=\mathrm{1,2},...,s\&times;s---\left(13\right)$

Key K (μ, x in step 9, the employing hidden image scramble process ₀), regenerate chaos sequence, then with the inverse process of scramble process, to the scramble that the extracts random conversion that is inverted that hides Info, recover the EW that hides Info.On the basis of traditional PCA Information Hiding Techniques, be the geometric attack of opposing image, the initial carrier image has been adopted image normalization technology based on geometric invariant moment, realize the geometry correction of image.

The described intensity factor that embeds of determining to hide Info of step 3 is adopting the PCA algorithm to ask between the Image Subspace, and employing is sheltered and texture masking based on the illumination of vision system, design perceptual mask template:

α＝c ₁·(1-NVF)+c ₂·NVF（10）

Determine the intensity factor that each sub-block hides Info and embeds.

From Fig. 4 (a) to 4(d) (Fig. 4 (a) is the initial carrier image; Fig. 4 (b) contains the image that hides Info, Fig. 4 (c) is the enciphering hiding information of extracting, what Fig. 4 (d) extracted hides Info) can find out, in the experiment of digital image hidden information, the information concealing method that combines based on image normalization and PCA has realized that the success of image hides, moreover ((Fig. 5 (a) is the initial carrier image to Fig. 5 (a) to Fig. 5 (l) from Fig. 5 (a) to Fig. 5 (l), Fig. 5 (b) hides Info 1, Fig. 5 (c) hides Info 2, Fig. 5 (d) is enciphering hiding information 1, Fig. 5 (e) is enciphering hiding information 2, Fig. 5 (g) contains the image that hides Info, and Fig. 5 (i) extracts enciphering hiding information 1, and Fig. 5 (j) extracts enciphering hiding information 2, Fig. 5 (k) extracts 1, the Fig. 5 (l) that hides Info to extract to hide Info 2.) it can also be seen that the many information concealing methods success that combines based on image normalization and PCA realization the embedding of a plurality of information, improved the embedding capacity of carrier image, have good robustness, geometric attack is also had good resistibility.

The above; only be the better embodiment of the present invention; but protection scope of the present invention is not limited to this; anyly be familiar with those skilled in the art in the technical scope that the present invention discloses; be equal to replacement or change according to technical scheme of the present invention and inventive concept thereof, all should be encompassed within protection scope of the present invention.

Claims (3)

1. many information concealing methods that combine based on image normalization and PCA is characterized in that: may further comprise the steps:

Step 1: be 256 grades of gray level image F={f (i, j) for initial carrier, 1≤i≤m, 1≤j≤m}, the image size is m * m, hiding Info is bianry image W={w (i, j), 1≤i≤s, 1≤j≤s}, the image size is s * s; Carrier image is carried out normalized based on geometric invariant moment, try to achieve normalized image I and image centroid thereof, and corresponding transformation parameter R;

Step 2, get and embed the zone and with its piecemeal;

Point centered by the barycenter of normalized image I at first is divided into piece image F N size and is the sub-block of s * s, each sub-block is rearranged the vector x that obtains n=s * s dimension _i, i=1,2 ..., N is zero in order to make its average, need to deduct its mean vector namely:

A＝(x ₁-μ ₀,x ₂-μ ₀,...,x _N-μ ₀)（1）

A is called the training sample matrix in the formula,

Be the population mean of sample, at this moment, the autocorrelation matrix R among the KLT has become covariance matrix, and its maximal possibility estimation is total population scatter matrix:

Because AA ^TAnd A ^TA has identical eigenwert, and AA ^TCorresponding to eigenvalue λ _iProper vector u _iWith A ^TThe corresponding proper vector v of A _iHave following relationship:

Then can be expressed as the sample xx among any X:

Y wherein _j=x _i ^Tu _j(j=1,2 ..., n), y=[y ₁, y ₂..., y _n] ^TChoose front d the component of y As feature, then can demonstrate,prove

Be d component of variance maximum (being that energy is maximum), with

Corresponding subspace is called as signal subspace, and then obtains sample vector A and total population scatter matrix S _t

Step 3, determine the intensity factor that hides Info

Design perceptual mask template is as follows:

α＝c ₁·(1-NVF)+c ₂·NVF（5）

Described c ₁And c ₂Be used for regulating the embedment strength that hides Info, c ₁Can be used to regulate the visual quality that contains the image that hides Info.Noise visible function (Noise visibility function, NVF) adopts following form:

Wherein, k adjusts parameter, and every width of cloth image has different adjustment parameters, The local variance of original image in the expression window;

The maximal value of expression original image local variance;

Step 4, take the initial parameter that generates the Logistic chaos sequence as encryption key, carry out Chaotic Scrambling to hiding Info, it comprises that step is as follows:

(1) input key K (μ, x ₀), produce chaos sequence X, X is arranged by ascending order, namely [X ', l]=sort (X), X' is the chaos sequence that ascending order is arranged, l is index sequence;

(2) W that will hide Info is converted to one-dimensional vector, then is rearranged to the scramble W' that hides Info with index sequence l;

Step 5, by formula (9) calculate S _tThe KL conversion coefficient, get its eigenwert y _i, the corresponding sub-space feature vectors u of i ∈ [1,2 .., N] _i, wherein, N is the sub-block number of cutting, then, and at u _iEmbed the scramble W' that hides Info, concrete formula is:

Its intensity factor α _jAccording to the local characteristics of different sub-blocks by formula (10) calculate;

Step 6, use amended sub-space feature vectors

Replace original u _i, carry out the sample vector after reconstruct is tried to achieve in the KL inverse transformation, add the average μ that deducts in the step 2 ₀, and be transformed to the image subblock of s * s, then be rearranged into and embed the area image that hides Info;

Step 7, hide Info area image and original image remainder of embedding is merged into and contains the carrier image that hides Info, carry out contrary normalization to containing the carrier image that hides Info by transformation parameter R, obtain to embed the carrier image F' after hiding Info, finish the embedding that hides Info;

The extraction of step 8, image information

Leaching process is as follows:

(1) utilization is treated detected image F' and is carried out normalized, to obtain corresponding normalized image I' based on the image normalization technology of square;

(2) with hide Info identically when embedding, equally the embedding of normalized image I' is hidden Info that extract in the zone and piecemeal, then be transformed to sample vector by formula (2) try to achieve its total population scatter matrix S _t';

KL transformation matrix when (3) utilization hides Info embedding is to S _t' carry out feature extraction, obtain corresponding to y _i, i ∈ [1,2 ..., N] sub-space feature vectors u _i';

Intensity factor when (4) embedding according to each image subblock, the setting threshold decision function embeds the extraction of the EW that hides Info, and formula is as follows:

Key K (μ, x in step 9, the employing hidden image scramble process ₀), regenerate chaos sequence, then with the inverse process of scramble process, to the scramble that the extracts random conversion that is inverted that hides Info, recover the EW that hides Info.

2. a kind of many information concealing methods that combine based on image normalization and PCA according to claim 1, it is characterized in that: on the basis of traditional PCA Information Hiding Techniques, geometric attack for the opposing image, the initial carrier image is adopted image normalization technology based on geometric invariant moment, realized the geometry correction of image.

3. a kind of many information concealing methods that combine based on image normalization and PCA according to claim 1, it is characterized in that: step 3 is described determines the intensity factor that hides Info and embed, it is characterized in that: before employing PCA algorithm is asked for Image Subspace, employing is sheltered and texture masking based on the illumination of vision system, design perceptual mask template: α=c ₁(1-NVF)+c ₂NVF determines the intensity factor that each sub-block hides Info and embeds.

Images