CN102129549A - Image Hash method based on thumbnail and singular value decomposition - Google Patents
Image Hash method based on thumbnail and singular value decomposition Download PDFInfo
- Publication number
- CN102129549A CN102129549A CN 201110033139 CN201110033139A CN102129549A CN 102129549 A CN102129549 A CN 102129549A CN 201110033139 CN201110033139 CN 201110033139 CN 201110033139 A CN201110033139 A CN 201110033139A CN 102129549 A CN102129549 A CN 102129549A
- Authority
- CN
- China
- Prior art keywords
- image
- thumbnail
- singular value
- feature
- hash
- 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
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention relates to an image Hash method based on thumbnail and singular value decomposition. The method comprises the following steps of: pre-processing an image and blocking; applying the singular value decomposition to an image block and extracting a first singular value and a second singular value as characteristics; generating the thumbnail of the original image by a bilinear interpolation method and performing the singular value decomposition, and taking the first singular value and the second singular value as the characteristics; normalizing the characteristics of the image block and the thumbnail, and rearranging the positions of the characteristic sequence of the image block; based on the first singular value and the second singular value of the thumbnail, calculating an Euclidean distance between each pair of singular values and the thumbnail, and connecting all the distance values in series, namely during similarity judgment of the image Hash, calculating an L1 norm of two image Hashes; if the L1 norm is smaller than a set threshold value, determining that the corresponding images are similar; and if the L1 norm is greater than or equal to the set threshold value, determining that the corresponding images are different. The method has robustness to common number processing such as joint photographic experts group (JPEG) compression, proper noise interference, brightness regulation, contrast enhancement and the like; and the method has high uniqueness.
Description
Technical field
The present invention relates to signal Processing field and field of computer technology, particularly a kind of image Hash method based on thumbnail and svd.
Background technology
Along with popularizing of image acquisition equipments such as digital camera, digital picture increases by geometric progression, and how effectively the management and retrieval digital picture becomes the problem that people face.Meanwhile, the characteristics that digital picture is easy to duplicate and revise make picture material infringement, distort, problem such as forgery is serious day by day, how to protect problems such as image copyright to become more and more important better.Image Hash (Hash) is a kind of emerging technology of multi-media information security.It identifies this image with the short sequence of extracting in the image, can be widely used in fields such as image authentication, copy detection, digital watermarking, image retrieval.
Because cryptography Hash function such as SHA-1 and MD5 are very responsive to the variation of input data, the change of any 1 bit all can change the hash value of output fully, therefore is not suitable for image.Often will carry out normal digital processing as enhancing, JPEG compression etc. to image in the practical application, material alterations does not take place in its content, wishes that image Hash remains unchanged.Usually, image Hash should satisfy two conditions: 1) perception robustness, and two promptly similar width of cloth images to perception, no matter whether interior data is consistent, their Hash is identical or very approaching with very big probability; 2) uniqueness, promptly different images has different image Hash.
According to the constructing technology difference, conventional images Hash method roughly is divided into following five classes: (1) is based on the method for image statistics; (2) based on the method for invariant relation; (3) based on the method for image rough representation; (4) utilize image low layer semantic feature; (5) utilization matrix decomposition technology.Specifically consult following document:
1.R.Venkatesan，S.-M.Koon，M.H.Jakubowski，et?al.，Robust?image?hashing[C]，in?Proc.ofthe?IEEE?International?Conference?on?Image?Processing，2000，3：664-666，Vancouver，BC，Canada，September?10-13，2000.
2.C.Y.Lin?and?S.F.Chang，A?robust?image?authentication?system?distinguishing?JPEGcompression?from?malicious?manipulation[J].IEEE?Transactions?on?Circuits?and?Systems?forVideo?Technology，2001，11(2)：153-168.
3.A.Swaminathan，Y.Mao?and?M.Wu，Robust?and?secure?image?hashing[J].IEEETransactions?on?Information?Forensics?and?Security，2006，1(2)：215-230.
4.V.Monga?and?B.L.Evans，Perceptual?image?hashing?via?feature?points：performanceevaluation?and?trade-offs[J].IEEE?Transactions?on?Image?Processing，2006，15(11)：3453-3466.
5.V.Monga?and?M.K.Mihcak，Robust?and?secure?image?hashing?via?non-negative?matrixfactorizations[J].IEEE?Transactions?on?Information?Forensics?and?Security，2007，2(3)：376-390.
Above-mentioned prior art has robustness to some digital processing mostly, as JPEG compression, digital filtering, geometric transformation, but the relatively poor deficiency of ubiquity uniqueness.
Summary of the invention
The object of the present invention is to provide a kind of image Hash method based on thumbnail and svd, this method can extract sane Hash from digital picture, sane to common Flame Image Process, and has good uniqueness.
A kind of image Hash method based on thumbnail and svd earlier to the input picture pre-service, is carried out non-overlapped piecemeal again; Then svd is applied to each image block, get the 1st and the 2nd singular value as block feature, convert to input picture with image block image of the same size and get luminance component and represent with bilinear interpolation, generate the thumbnail of original image, simultaneously svd is applied to thumbnail, get the 1st and the 2nd singular value as feature, characteristic value normalization with image block and thumbnail, and the characteristic sequence of image block carried out position rearrangement reaction, with the 1st of thumbnail, 2 singular values are benchmark, calculate all the other every pair singular value and its Euclidean distance, all distance values of contacting are image Hash, when judging two image Hash similaritys, calculate their L1 norm, if less than setting threshold, think that its corresponding image is identical, otherwise think different images.
The concrete steps of this method are as follows:
(1) image pre-service: input picture I is carried out pre-service, comprise picture size normalization and color space conversion.Earlier use bilinear interpolation, picture specification is changed into M * M size,, then it is transformed into the YCbCr space, get Y component representative image if be input as coloured image.Remember that pretreated image is J;
(2) image block: J is divided into size is the non overlapping blocks of t * t, a total N=M
2/ t
2Individual image block (image is returned when formatting, and gets the integral multiple that M is t) is numbered by from left to right inferior ordered pair piecemeal from top to bottom, remembers that i image block is B
i(1≤i≤N);
(3) computed image block feature: establish B
iSvd be designated as [U
iS
iV
i]=SVD (B
i), U wherein
iAnd V
iBe unitary matrix, S
iBe diagonal matrix, S
iElement on the diagonal line is B
iSingular value, get B
iThe the 1st, 2 singular value as the image block feature, be designated as p respectively
iAnd q
i, i.e. p
i=S
i(1,1) and q
i=S
i(2,2).Contact respectively the 1st, 2 feature of image block can obtain describing the proper vector p=[p of entire image
1, p
2..., p
N] and q=[q
1, q
2..., q
N];
(4) generate thumbnail: utilize bilinear interpolation, image I is normalized into size for the thumbnail of t * t,, further it is transformed into the YCbCr space, get Y component representative image, remember that final thumbnail is R for coloured image;
(5) extract thumbnail feature: R is carried out svd, get its 1st, 2 singular value, be designated as s respectively as feature
1And s
2
(6) feature normalization: the 1st, 2 singular value, i.e. u using vectorial u and v presentation video piece and thumbnail respectively
i=p
i(1≤i≤N), u
N+1=s
1v
i=q
i(1≤i≤N), v
N+1=s
2, respectively to the element normalization of u and v, i.e. x
i=(u
i-μ
u)/δ
u, y
i=(v
i-μ
v)/δ
v, μ wherein
uAnd μ
vBe respectively the average of u and v, δ
uAnd δ
vBe respectively their standard deviation;
(7) feature scramble: under cipher controlled, respectively the top n element of normalized vector x and y is carried out position rearrangement reaction with pseudo-random generator, and it is constant to keep N+1 element, obtains vector x behind the scramble ' and y ';
(8) Hash extracts, and calculates
(1≤i≤N), can obtain image Hash h=[h
1, h
2..., h
N];
(9) similarity is judged, establishes h
(1)And h
(2)Be respectively two image Hash sequences, h
i (1)And h
i (2)I element representing them respectively calculates the L1 norm
If d, thinks h less than setting threshold T
(1)And h
(2)Pairing image is identical, otherwise thinks different images.
The present invention compared with prior art has following conspicuous outstanding substantive distinguishing features and marked improvement: the present invention uses svd to extract image information, gets most important two singular values of image block as feature; Introduce image thumbnails as a reference, the distance of using image block singular value and thumbnail singular value is as hash value; The image Hash that extracts is sane to common digital processings such as JPEG compression, comfort noise interference, brightness adjustment, contrast enhancings, and has good uniqueness.
Description of drawings
Fig. 1 is a width of cloth test pattern of using in the embodiment of the invention, and size is 600 * 400;
Fig. 2 is the other piece image of using in the embodiment of the invention;
Fig. 3 carries out the result that size block is formatted to Fig. 1;
Fig. 4 is Y component (luminance component) result who gets Fig. 3, also promptly Fig. 1 is carried out pretreated net result;
Fig. 5 is the piecemeal synoptic diagram of Fig. 4;
Fig. 6 is the thumbnail of Fig. 1;
Fig. 7 is the statistical Butut of L1 norm among the embodiment.
Embodiment
Below in conjunction with accompanying drawing a preferred embodiment of the present invention is elaborated, but protection scope of the present invention is not limited to following embodiment.
Embodiment:
Present embodiment comprises robustness checking and uniqueness checking two parts.Robustness checking is by judging whether similar realization of the pairing image Hash of Fig. 1 and Fig. 2, wherein Fig. 2 obtains by Fig. 1 is carried out continuous digital processing, comprises that JPEG compression (quality factor is 60), brightness adjustment (adjusting range is 20), contrast strengthen (adjusting range is 20) and white Gaussian noise (average is 0, variance be 0.01).In the step below, (1)～(8) are the steps of extracting the Hash of Fig. 1, extract step and Fig. 1 identical of the Hash of Fig. 2, repeated description no longer, and (9) are the similarity judgement of two image Hash, concrete steps are as follows:
(1) image pre-service: with bilinear interpolation Fig. 1 is normalized into 256 * 256 sizes, the result as shown in Figure 3; Fig. 3 is transformed into the YCbCr space representation, gets Y component representative image, obtain pretreated result, as shown in Figure 4;
(2) image block: it is 64 * 64 non overlapping blocks that Fig. 4 is divided into size, obtains 16 image blocks altogether, and Fig. 5 is the piecemeal synoptic diagram;
(3) computed image block feature: above-mentioned 16 image blocks are carried out svd successively, and the 1st, 2 singular value of getting each image block is as feature, so can obtain the proper vector p and the q of image:
p＝[13618.85，12438.3281，6800.47315，5791.30954，13182.7201，12360.2039，7554.88336，7234.64，12772.461，11346.9951，6586.04545，7960.01579，8250.24471，5453.8304，5120.26638，5752.40614]，
q＝[130.76731，623.10121，484.34539，381.7512，112.65023，516.03795，665.47482，606.93448，193.00764，971.94397，844.79315，288.13595，1315.9135，163.94719，247.47046，221.32931]
(4) generate thumbnail: utilizing bilinear interpolation that Fig. 1 is normalized into size is 64 * 64, is transformed into the YCbCr space simultaneously, represents with the Y component, obtains thumbnail as shown in Figure 6;
(5) extract thumbnail feature: Fig. 6 is carried out svd, obtain the 1st, 2 singular value, i.e. s
1=9323.15671, s
2=130.76731;
(6) feature normalization: with the 1st, 2 singular value of vectorial u and v presentation video piece and thumbnail, promptly
u＝[13618.85，12438.3281，6800.47315，5791.30954，13182.7201，12360.2039，7554.88336，7234.64，12772.461，11346.9951，6586.04545，7960.01579，8250.24471，5453.8304，5120.26638，5752.40614，9323.15671]，
v＝[130.76731，623.10121，484.34539，381.7512，112.65023，516.03795，665.47482，606.93448，193.00764，971.94397，844.79315，288.13595，1315.9135，163.94719，247.47046，221.32931，130.76731]
U and v are carried out normalization obtain vector x and y:
x＝[1.5486，1.16，-0.69593，-1.0281，1.4051，1.1343，-0.44758，-0.553，1.27，0.80075，-0.76651，-0.31421，-0.21867，-1.1392，-1.249，-1.0409，0.13452]，
y＝[-0.97634，0.46352，0.057717，-0.24232，-1.0293，0.1504，0.58744，0.41624，-0.79432，1.4837，1.1119，-0.51611，2.4897，-0.8793，-0.63504，-0.71149，-0.97634]
(7) feature scramble: preceding 16 elements to normalized vector x and y carry out position rearrangement reaction, and it is constant to keep the 17th element, obtain vector x behind the scramble ' and y ':
x′＝[-0.69593，1.16，-0.31421，-1.249，-1.0409，-0.76651，1.5486，0.80075，1.4051，-0.21867，-1.0281，-0.44758，-1.1392，-0.553，1.1343，1.27，0.13452]，
y′＝[0.71149，-1.0293，1.1119，-0.24232，-0.51611，0.58744，-0.79432，2.4897，1.4837，-0.63504，-0.97634，0.41624，0.1504，-0.8793，0.057717，0.46352，-0.97634]
(8) Hash extracts, and calculates
(1≤i≤16) can obtain the image Hash:h of Fig. 1
(1)=[0.87166,1.0269,2.1359,1.5662,1.2623,1.8048,1.4258,3.5295,2.7688,0.49116,1.1627,1.5093,1.7006,0.69434,1.4383,1.8337].
(9) similarity is judged, the image Hash that extracts Fig. 2 obtains h
(2)=[0.9612,1.1092,2.3191,1.7667,1.4140,2.0047,1.1778,3.5129,2.8024,0.7206,1.1775,1.7879,1.8365,0.7194,1.6644,1.9184]; Setting threshold T=1.0 calculates h
(1)And h
(2)The L1 norm, obtain d=0.64987; Because d less than T, therefore can think that Fig. 1 and Fig. 2 are similar images.
The present invention with 100 width of cloth different images (size is 256 * 256～1994 * 2592) as test data, the image Hash that extracts them (gets M=256, t=64) and calculate the L1 norm between Hash in twos, have 4950 results, its statistical Butut as shown in Figure 7.Wherein, minor increment is 1.6458, and ultimate range is 7.5218, is 4.3755 apart from average, and standard deviation is 0.8587.Find from embodiment, the Hash of any two width of cloth different images, its distance illustrates that all greater than setting threshold T=1.0 the present invention has uniqueness preferably.
Claims (2)
1. image Hash method based on thumbnail and svd, it is characterized in that: earlier to the input picture pre-service, carry out non-overlapped piecemeal again, then svd is applied to each image block, get the 1st and the 2nd singular value as block feature, convert to input picture with image block image of the same size and get luminance component and represent with bilinear interpolation, generate the thumbnail of original image, simultaneously svd is applied to thumbnail, get the 1st and the 2nd singular value as feature, characteristic value normalization with image block and thumbnail, and the characteristic sequence of image block carried out position rearrangement reaction, with the 1st of thumbnail, 2 singular values are benchmark, calculate every pair of singular value and its Euclidean distance, all distance values of contacting are image Hash, when judging two image Hash similaritys, calculate their L1 norm, if less than setting threshold, think the image similarity that it is corresponding, otherwise think different images.
2. the image Hash method based on thumbnail and svd according to claim 1, it is characterized in that: concrete steps are as follows:
(1) image pre-service: input picture I is carried out pre-service, comprise picture size normalization and color space conversion.Adopt bilinear interpolation earlier, picture specification is changed into M * M size; If be input as coloured image, then it is transformed into the YCbCr space, with Y component representative image, remember that pretreated image is J;
(2) image block: J is divided into size is the non overlapping blocks of t * t, a total N=M
2/ t
2Individual image block, image are returned when formatting, and get the integral multiple that M is t, number by from left to right inferior ordered pair piecemeal from top to bottom, remember that i image block is Bi (1≤i≤N);
(3) computed image block feature: establish B
iSvd be designated as [U
iS
iV
i]=SVD (B
i), U wherein
iAnd V
iBe unitary matrix, S
iBe diagonal matrix, S
iElement on the diagonal line is B
iSingular value, get B
iThe the 1st, 2 singular value as the image block feature, be designated as p respectively
iAnd q
i, i.e. p
i=S
i(1,1) and q
i=S
i(2,2), the 1st, 2 feature of the image block of contacting respectively can obtain describing the proper vector p=[p of entire image
1, p
2..., p
N] and q=[q
1, q
2..., q
N];
(4) generate thumbnail: utilize bilinear interpolation, image I is normalized into size for the thumbnail of t * t,, it is transformed into the YCbCr space, get Y component representative image, remember that final thumbnail is R for coloured image;
(5) extract thumbnail feature: R is carried out svd, get its 1st, 2 singular value, be designated as s respectively as feature
1And s
2
(6) feature normalization: use the feature of vectorial u and v presentation video piece and image thumbnails respectively, i.e. u
i=p
i(1≤i≤N), u
N+1=s
1v
i=q
i(1≤i≤N), v
N+1=s
2, the element to u and v carries out normalization, i.e. x respectively
i=(u
i-μ
u)/δ
u, y
i=(v
i-μ
v)/δ
v, wherein, μ
uAnd μ
vBe respectively the average of u and v, δ
uAnd δ
vBe respectively their standard deviation;
(7) feature scramble: under cipher controlled, respectively the top n element of normalized vector x and y is carried out position rearrangement reaction with pseudo-random generator, and it is constant to keep N+1 element, obtains vector x behind the scramble ' and y ';
(8) Hash extracts: calculate
(1≤i≤N), can obtain image Hash h=[h
1, h
2..., h
N];
(9) similarity is judged: establish h
(1)And h
(2)Be respectively two image Hash sequences,
With
I element representing them respectively, computed range
If d, thinks h less than setting threshold T
(1)And h
(2)Pairing image similarity, otherwise think different images.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110033139A CN102129549B (en) | 2011-01-29 | 2011-01-29 | Image Hash method based on thumbnail and singular value decomposition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110033139A CN102129549B (en) | 2011-01-29 | 2011-01-29 | Image Hash method based on thumbnail and singular value decomposition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102129549A true CN102129549A (en) | 2011-07-20 |
| CN102129549B CN102129549B (en) | 2012-09-05 |
Family
ID=44267629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110033139A Expired - Fee Related CN102129549B (en) | 2011-01-29 | 2011-01-29 | Image Hash method based on thumbnail and singular value decomposition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102129549B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102881008A (en) * | 2012-08-16 | 2013-01-16 | 广西师范大学 | Circular loop statistic characteristic-based anti-rotation image Hash method |
| CN104778689A (en) * | 2015-03-30 | 2015-07-15 | 广西师范大学 | Image digest method based on mean secondary image and locality preserving projection |
| CN106179992A (en) * | 2016-08-26 | 2016-12-07 | 保定市立中车轮制造有限公司 | A kind of wheel hub Automated Sorting System and method for sorting thereof |
| CN109598726A (en) * | 2018-10-26 | 2019-04-09 | 哈尔滨理工大学 | A kind of adapting to image target area dividing method based on SLIC |
| CN109685112A (en) * | 2018-11-29 | 2019-04-26 | 昆明理工大学 | It is a kind of based on color difference algorithm determination method similar with the image of DHash |
| CN114240827A (en) * | 2021-11-04 | 2022-03-25 | 山东师范大学 | Half-parameter sensing encryption visual security analysis method and system based on hash |
| CN118657799A (en) * | 2024-08-16 | 2024-09-17 | 武汉泰铭恒创信息技术股份有限公司 | Traffic police intelligent management method and system based on big data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101030293A (en) * | 2007-03-30 | 2007-09-05 | 西安电子科技大学 | Digital watermark method against geometrical attack based on image characteristic region |
| WO2009046438A1 (en) * | 2007-10-05 | 2009-04-09 | Dolby Laboratories Licensing Corp. | Media fingerprints that reliably correspond to media content |
| CN101702230A (en) * | 2009-11-10 | 2010-05-05 | 大连理工大学 | Stable digital watermark method based on feature points |
| CN101710334A (en) * | 2009-12-04 | 2010-05-19 | 大连理工大学 | Large-scale image library retrieving method based on image Hash |
-
2011
- 2011-01-29 CN CN201110033139A patent/CN102129549B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101030293A (en) * | 2007-03-30 | 2007-09-05 | 西安电子科技大学 | Digital watermark method against geometrical attack based on image characteristic region |
| WO2009046438A1 (en) * | 2007-10-05 | 2009-04-09 | Dolby Laboratories Licensing Corp. | Media fingerprints that reliably correspond to media content |
| CN101702230A (en) * | 2009-11-10 | 2010-05-05 | 大连理工大学 | Stable digital watermark method based on feature points |
| CN101710334A (en) * | 2009-12-04 | 2010-05-19 | 大连理工大学 | Large-scale image library retrieving method based on image Hash |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102881008A (en) * | 2012-08-16 | 2013-01-16 | 广西师范大学 | Circular loop statistic characteristic-based anti-rotation image Hash method |
| CN102881008B (en) * | 2012-08-16 | 2016-06-01 | 广西师范大学 | Based on the anti-rotation image Hash method of annulus statistical nature |
| CN104778689A (en) * | 2015-03-30 | 2015-07-15 | 广西师范大学 | Image digest method based on mean secondary image and locality preserving projection |
| CN104778689B (en) * | 2015-03-30 | 2018-01-05 | 广西师范大学 | A kind of image hashing method based on average secondary image and locality preserving projections |
| CN106179992A (en) * | 2016-08-26 | 2016-12-07 | 保定市立中车轮制造有限公司 | A kind of wheel hub Automated Sorting System and method for sorting thereof |
| CN106179992B (en) * | 2016-08-26 | 2018-04-13 | 保定市立中车轮制造有限公司 | A kind of wheel hub Automated Sorting System and its method for sorting |
| CN109598726A (en) * | 2018-10-26 | 2019-04-09 | 哈尔滨理工大学 | A kind of adapting to image target area dividing method based on SLIC |
| CN109685112A (en) * | 2018-11-29 | 2019-04-26 | 昆明理工大学 | It is a kind of based on color difference algorithm determination method similar with the image of DHash |
| CN114240827A (en) * | 2021-11-04 | 2022-03-25 | 山东师范大学 | Half-parameter sensing encryption visual security analysis method and system based on hash |
| CN118657799A (en) * | 2024-08-16 | 2024-09-17 | 武汉泰铭恒创信息技术股份有限公司 | Traffic police intelligent management method and system based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102129549B (en) | 2012-09-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102129549B (en) | Image Hash method based on thumbnail and singular value decomposition | |
| Mushtaq et al. | Digital image forgeries and passive image authentication techniques: a survey | |
| Tang et al. | Robust image hash function using local color features | |
| Sajjad et al. | Robust image hashing based efficient authentication for smart industrial environment | |
| Armas Vega et al. | Copy-move forgery detection technique based on discrete cosine transform blocks features | |
| Hakimi et al. | Image splicing forgery detection using local binary pattern and discrete wavelet transform | |
| CN103238159A (en) | System and method for image authentication | |
| Sharma et al. | Comprehensive analyses of image forgery detection methods from traditional to deep learning approaches: an evaluation | |
| Qu et al. | A convolutive mixing model for shifted double JPEG compression with application to passive image authentication | |
| Bharti et al. | A survey of image forgery detection techniques | |
| Murali et al. | Comparision and analysis of photo image forgery detection techniques | |
| EP2149098A1 (en) | Deriving video signatures that are insensitive to picture modification and frame-rate conversion | |
| Gharde et al. | Robust perceptual image hashing using fuzzy color histogram | |
| Tang et al. | Perceptual image hashing with histogram of color vector angles | |
| Gupta et al. | Video authentication in digital forensic | |
| Hong et al. | A color image authentication scheme with grayscale invariance | |
| Saikia et al. | Image authentication under geometric attacks via concentric square partition based image hashing | |
| Zhang et al. | Image splicing detection scheme based on error level analysis and local binary pattern | |
| Das et al. | Blind detection method for video inpainting forgery | |
| Sandeep et al. | Detection of malicious video modifications using perceptual video hashing | |
| CN102881008A (en) | Circular loop statistic characteristic-based anti-rotation image Hash method | |
| Govindaraj et al. | Ring partition and DWT based perceptual image hashing with application to indexing and retrieval of near-identical images | |
| Ustubıoglu et al. | Image forgery detection using colour moments | |
| Rahmani et al. | High hiding capacity steganography method based on pixel indicator technique | |
| Neelima et al. | Perceptual hash function for images based on hierarchical ordinal pattern |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120905 Termination date: 20190129 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |