CN106815852A - Coloured image evidence collecting method based on the extremely humorous conversion of quaternary number - Google Patents
Coloured image evidence collecting method based on the extremely humorous conversion of quaternary number Download PDFInfo
- Publication number
- CN106815852A CN106815852A CN201611210320.5A CN201611210320A CN106815852A CN 106815852 A CN106815852 A CN 106815852A CN 201611210320 A CN201611210320 A CN 201611210320A CN 106815852 A CN106815852 A CN 106815852A
- Authority
- CN
- China
- Prior art keywords
- image
- feature
- block
- region
- blocks
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000006243 chemical reaction Methods 0.000 title abstract 3
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 238000005070 sampling Methods 0.000 claims abstract description 4
- 239000013598 vector Substances 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 12
- 230000009466 transformation Effects 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 4
- 230000006978 adaptation Effects 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims description 3
- 230000000877 morphologic effect Effects 0.000 claims description 3
- 238000012805 post-processing Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 28
- 238000012360 testing method Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 7
- 238000002474 experimental method Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 241001255830 Thema Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of coloured image evidence collecting method based on the extremely humorous conversion of quaternary number, coloured image to be detected is carried out into the adaptivenon-uniform sampling based on entropy rate super-pixel first, join probability density SURF operator extractions characteristic point simultaneously determines local suspicious tampered region；Then, sliding window operation is carried out to suspicious region, represents that the feature of each sliding window block carries out similarity mode using the extremely humorous conversion coefficient square value of quaternary number；Finally, it is determined that and marking tampered region.Test result indicate that, compared to altering detecting method of the tradition based on piecemeal, the method for the present invention dramatically reduces the time complexity of algorithm while detection accuracy higher is kept.
Description
Technical Field
The invention relates to a method for detecting image tampering of copying and pasting, in particular to a color image evidence obtaining method based on quaternion polar harmonic transformation, and belongs to the technical field of digital image authentication.
Background
In recent years, with the rapid development of network and multimedia technologies, digital multimedia resources are more and more susceptible to illegal tampering and counterfeiting, and the legal rights and interests of copyright owners are seriously infringed. Therefore, multimedia security closely related to the security is receiving more and more attention, and effective protection of intellectual property is an important research topic. The image authentication technology is a method for identifying the integrity, trueness and other attributes of a digital image, and the tampering detection technology is one of the commonly used methods.
At present, the digital image authentication technology is developed rapidly, the tamper detection technology is a very important branch, and the existing tamper detection technology is mainly divided into active detection and passive detection. The passive detection can also be called blind detection, the technology does not need to rely on the original information of the image to be detected, the target image can be directly subjected to tampering detection, the utilization value and the practical value are higher, and the technology is a main research trend in the current image authentication technology. Tampering with an image generally leaves no visually recognizable clues, but the intrinsic characteristics of the image always change more or less under tampering, which is the most important basis for blind detection techniques.
Most image tampering methods, which alter image content and statistical characteristics, are done from a pixel level. In view of such tampering methods, researchers have proposed a series of targeted detection algorithms, for example, a detection algorithm for copy-paste tampering and a detection algorithm for splice tampering, and some algorithms perform tampering detection by using changes in statistical characteristics before and after image tampering. The detection method for copy-paste tampering can be further summarized as a block-based detection method and a feature point-based detection method: the block-based method generally has strong robustness, the detection precision is relatively high, but the time complexity seriously exceeds the acceptance range of people; although the problem of time complexity is well solved, the detection accuracy of the method based on the feature points is far lower than that of the method based on the blocks.
Disclosure of Invention
The invention aims to solve the technical problems of the existing copy-paste tamper detection technology and provides a color image evidence obtaining method based on quaternion polar harmonic transformation, which can process random copy-paste.
The technical solution of the invention is as follows: a color image evidence obtaining method based on quaternion polar harmonic transformation is characterized by comprising the following steps:
appointing:indicating a color image to be detected;respectively representing the number of rows and the number of columns of the image;the image is a smooth image after preprocessing;initializing the number of superpixels for self adaptation;a local feature region of the structure;matching a threshold value for the feature point; matrix arrayUsed for storing the correlation coefficient between the blocks;matching a threshold for a block; matrix arrayIs a set of characteristic vectors;
a. initial setting
Acquiring an image to be detected and initializing variables;
b. self-adaptive superpixel block of image to be detected
b.1 imagingPerforming a smoothing pretreatment, andis convoluted to obtain，Wherein:
；
b.2 willFour-level non-down sampling shear wave transformation is carried out to obtain the normalized low-frequency coefficientAnd normalized high frequency coefficient；
b.3 calculationLow frequency energyHigh frequency energyAndin proportion of total energy；
b.4 calculating the number of adaptive superpixels according to the following formula：
；
b.5 integration Using entropy Rate superpixel segmentation AlgorithmSegmenting an image to be detected;
c. probability density SURF feature point extraction structure
c.1 calculationProbability density of each point in the luminance component;
c.2 calculating the second derivative of the probability density of each point according to the following formula:
；
c.3 obtaining a second order autocorrelation matrix of points using an improved method；
c.4 construct SURF feature point detector:
c.5 useMaximum eigenvalue ofMinimum eigenvalueAnd feature vectorsCalculating the major semi-axisShort half shaftAnd angle of directionTo construct the pointThe elliptical local feature area of (a);
c.6 map the elliptical region pixels to the corresponding circular regions:
wherein,、、for the parameters required for the three mappings,is the position of the center of a circle of the circular area,andrespectively representing circular region coordinates obtained through mapping;
c.7 each local circular area is "filled with 0" to obtain a circumscribed square image, i.e. the image；
d.Low-order quaternion PHT feature expression and matching
d.1 is calculated according toQuaternion PHT decomposition of the region moment:
wherein:andrepresenting the R, G, B components of the color image in a polar coordinate system,anda conventional PHT representing R, G, B components;
d.2 finding eachThe 12 quaternion PHT moments represent the feature vectors of the SURF feature points, and all the SURF feature points and the vectors thereof in each superpixel block collectively represent the block features, i.e. the block features；
d.3 useCalculating the number of matched characteristic points in any two super-pixel blocks as the similarity coefficient of the two blocks；
d.4 storing inter-block correlation coefficients in ascending orderIn (1),wherein；
d.5 calculating separatelyFirst derivative ofSecond derivative ofAnd first derivative mean(ii) a In a matrixIs selected to satisfyAnd the value of the coefficient having the smallest value is taken as；
d.6 if the similarity coefficient of two super-pixel blocks is greater thanThe two blocks are regarded as matching blocks, namely the suspected tampering region SR;
e. superpixel sliding window method for determining and marking tampered area
e.1 dividing SR into areas by sliding Window operationEach having a radius of；
e.2 computing the quaternion PHT moment value of each circular blockSelecting 12 of themAs a stored set of feature vectorsPerforming the following steps;
e.3 based onAlgorithm, pairPerforming similar image block feature matching;
e.4 post-processing operations using RANSAC algorithm;
e.5 morphological operations are performed on the final defined area to fill in holes and remove individual blocks to mark tampered areas.
Firstly, carrying out self-adaptive segmentation on a color image to be detected based on entropy rate superpixels, extracting feature points by combining with a probability density SURF operator and determining a local suspicious tampered region; then, carrying out sliding window operation on the suspicious region, and carrying out similarity matching on the characteristics of each sliding window block by using the moment value of the quaternion polar harmonic transformation coefficient; finally, a tampered area is determined and marked. The experimental result shows that compared with the traditional block-based tampering detection method, the method provided by the invention has the advantages that the time complexity of the algorithm is greatly reduced while the higher detection accuracy is kept.
Compared with the prior art, the invention has the following beneficial effects:
firstly, because a large amount of time is consumed in the characteristic stage and the characteristic matching stage of each block in the sliding window operation of the traditional method, the time consumption of the prior art exceeds an ideal range, and compared with the sliding window block division of the whole image, the method effectively reduces the proportion of required sliding window pixel points, greatly reduces the time complexity of the algorithm, and has stronger practicability;
secondly, the SURF feature point extraction and probability density combination method enables the method to have stronger stability, enables the feature point distribution to be more uniform, and enables the feature region construction stage to be more effective;
and thirdly, the quaternion PHT has the characteristics of remarkable geometric invariance and the like, so that the method has stronger robustness, can be ideally detected under the condition that geometric transformation such as RST and the like is carried out on a tampered region, ensures the calculation efficiency by using low-order calculation, and does not increase the extra calculation time.
Drawings
Fig. 1 is an original diagram and a tampered diagram according to an embodiment of the present invention.
Fig. 2 is a diagram of intermediate results of various stages of a tamper detection process according to an embodiment of the present invention.
FIG. 3 is a diagram showing the results of the detection of the FAU library part according to the embodiment of the present invention.
FIG. 4 is a diagram of the detection results of GRIP library parts according to the embodiment of the present invention.
FIG. 5 is a flow chart of an embodiment of the present invention.
Detailed Description
The method comprises four stages in total: self-adaptive superpixel blocking and probability density SURF feature point extraction and feature region of image to be detectedThe structure,The method for expressing and matching the low-order quaternion polar harmonic transformation characteristics and determining and marking the tampered area by the superpixel sliding window method.
The specific steps are shown in fig. 5:
appointing:indicating a color image to be detected;respectively representing the number of rows and the number of columns of the image;the image is a smooth image after preprocessing;initializing the number of superpixels for self adaptation;a local feature region of the structure;matching a threshold value for the feature point; matrix arrayUsed for storing the correlation coefficient between the blocks;matching a threshold for a block; matrix arrayIs a set of characteristic vectors;
a. initial setting
Acquiring an image to be detected and initializing variables;
b. self-adaptive superpixel block of image to be detected
b.1 imagingPerforming a smoothing pretreatment, andis convoluted to obtain，Wherein:
；
b.2 willFour-level non-down sampling shear wave transformation is carried out to obtain the normalized low-frequency coefficientAnd normalized high frequency coefficient；
b.3 calculating Low frequency energyHigh frequency energyAndin proportion of total energy；
b.4 calculating the number of adaptive superpixels according to the following formula：
；
b.5 integration Using entropy Rate superpixel segmentation AlgorithmSegmenting an image to be detected;
c. probability density SURF feature point extraction structure
c.1 calculationProbability density of each point in the luminance component;
c.2 calculating the second derivative of the probability density of each point according to the following formula:
；
c.3 obtaining a second order autocorrelation matrix of points using an improved method；
c.4 construct SURF feature point detector:
c.5 useMaximum eigenvalue ofMinimum eigenvalueAnd (c) aEigenvectorCalculating the major semi-axisShort half shaftAnd angle of directionTo construct the pointThe elliptical local feature area of (a);
c.6 map the elliptical region pixels to the corresponding circular regions:
wherein,、、for the parameters required for the three mappings,is the position of the center of a circle of the circular area,andrespectively representing the mapped circular regionMarking;
c.7 each local circular area is "filled with 0" to obtain a circumscribed square image, i.e. the image；
d.Low-order quaternion PHT feature expression and matching
d.1 is calculated according toQuaternion PHT decomposition of the region moment:
wherein:andrepresenting the R, G, B components of the color image in a polar coordinate system,anda conventional PHT representing R, G, B components;
d.2 finding eachThe 12 quaternion PHT moments represent the feature vectors of the SURF feature points, and all the SURF feature points and the vectors thereof in each superpixel block collectively represent the block features, i.e. the block features；
d.3 useCalculating the number of matched characteristic points in any two super-pixel blocks as the similarity coefficient of the two blocks；
d.4 storing inter-block correlation coefficients in ascending orderIn (1),wherein；
d.5 calculating separatelyFirst derivative ofSecond derivative ofAnd first derivative mean(ii) a In a matrixIs selected to satisfyAnd the value of the coefficient having the smallest value is taken as；
d.6 if the similarity coefficient of two super-pixel blocks is greater thanThe two blocks are regarded as matching blocks, namely the suspected tampering region SR;
e. superpixel sliding window method for determining and marking tampered area
e.1 dividing SR into areas by sliding Window operationEach having a radius of；
e.2 computing the quaternion PHT moment value of each circular blockSelecting 12 of them as feature vector storage setPerforming the following steps;
e.3 based onAlgorithm, pairPerforming similar image block feature matching;
e.4 post-processing operations using RANSAC algorithm;
e.5 morphological operations are performed on the final defined area to fill in holes and remove individual blocks to mark tampered areas.
Experimental testing and parameter setting:
the experiment is performed in Matlab R2011a environment, the image involved in the experiment is an image in the FAU and GRIP image library common to most algorithms, the image size in the FAU image library is relatively large, the largest size image exceeds 3000 × 2400 pixels, and the area of the tampered region in the tampered image exceeds 6% of the total area. In order to improve the efficiency of the test process, a GRIP image library is newly introduced into the part, the size of images in the GRIP image library is 768 multiplied by 1024 pixels, and the size of a tampered area in the images is 4000 pixels to 50000 pixels.
An original graph and a tampered graph of an embodiment of the invention are shown in fig. 1.
An intermediate result diagram of each stage of the tampering detection process in the embodiment of the invention is shown in fig. 2.
The detection result diagram of the FAU library part in the embodiment of the invention is shown in FIG. 3.
Fig. 4 shows a graph of the detection results of the GRIP library according to the embodiment of the present invention.
Claims (1)
1. A color image evidence obtaining method based on quaternion polar harmonic transformation is characterized by comprising the following steps:
appointing:indicating a color image to be detected;respectively representing the number of rows and the number of columns of the image;the image is a smooth image after preprocessing;initializing the number of superpixels for self adaptation;a local feature region of the structure;matching a threshold value for the feature point; matrix arrayUsed for storing the correlation coefficient between the blocks;matching a threshold for a block; matrix arrayIs a set of characteristic vectors;
a. initial setting
Acquiring an image to be detected and initializing variables;
b. self-adaptive superpixel block of image to be detected
b.1 imagingPerforming a smoothing pretreatment, andis convoluted to obtain，Wherein:
；
b.2 willFour-level non-down sampling shear wave transformation is carried out to obtain the normalized low-frequency coefficientAnd normalized high frequency coefficient；
b.3 calculating Low frequency energyHigh frequency energyAndin proportion of total energy；
b.4 calculating the number of adaptive superpixels according to the following formula：
；
b.5 integration Using entropy Rate superpixel segmentation AlgorithmSegmenting an image to be detected;
c. probability density SURF feature point extraction structure
c.1 calculationProbability density of each point in the luminance component;
c.2 calculating the second derivative of the probability density of each point according to the following formula:
；
c.3 obtaining a second order autocorrelation matrix of points using an improved method；
c.4 construct SURF feature point detector:
c.5 useMaximum eigenvalue ofMinimum eigenvalueAnd feature vectorsCalculating the major semi-axisShort half shaftAnd angle of directionTo construct the pointThe elliptical local feature area of (a);
c.6 map the elliptical region pixels to the corresponding circular regions:
wherein,、、for the parameters required for the three mappings,is the position of the center of a circle of the circular area,andrespectively representing circular region coordinates obtained through mapping;
c.7 each local circular area is "filled with 0" to obtain a circumscribed square image, i.e. the image；
d.Low-order quaternion PHT feature expression and matching
d.1 is calculated according toQuaternion PHT decomposition of the region moment:
wherein:andrepresenting the R, G, B components of the color image in a polar coordinate system,anda conventional PHT representing R, G, B components;
d.2 finding eachThe 12 quaternion PHT moments represent the feature vectors of the SURF feature points, and all the SURF feature points and the vectors thereof in each superpixel block collectively represent the block features, i.e. the block features；
d.3 useCalculating the number of matched feature points in any two super-pixel blocks as the numberSimilarity coefficient of two blocks；
d.4 storing inter-block correlation coefficients in ascending orderIn (1),wherein；
d.5 calculating separatelyFirst derivative ofSecond derivative ofAnd first derivative mean(ii) a In a matrixIs selected to satisfyAnd the value of the coefficient having the smallest value is taken as；
d.6 if the similarity coefficient of two super-pixel blocks is greater thanThe two blocks are regarded as matching blocks, namely the suspected tampering region SR;
e. superpixel sliding window method for determining and marking tampered area
e.1 dividing SR into areas by sliding Window operationEach having a radius of；
e.2 computing the quaternion PHT moment value of each circular blockSelecting 12 of them as feature vector storage setPerforming the following steps;
e.3 based onAlgorithm, pairPerforming similar image block feature matching;
e.4 post-processing operations using RANSAC algorithm;
e.5 morphological operations are performed on the final defined area to fill in holes and remove individual blocks to mark tampered areas.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611210320.5A CN106815852B (en) | 2016-12-24 | 2016-12-24 | Color image evidence collecting method based on the extremely humorous transformation of quaternary number |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611210320.5A CN106815852B (en) | 2016-12-24 | 2016-12-24 | Color image evidence collecting method based on the extremely humorous transformation of quaternary number |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106815852A true CN106815852A (en) | 2017-06-09 |
| CN106815852B CN106815852B (en) | 2019-08-27 |
Family
ID=59110093
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611210320.5A Expired - Fee Related CN106815852B (en) | 2016-12-24 | 2016-12-24 | Color image evidence collecting method based on the extremely humorous transformation of quaternary number |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106815852B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109919927A (en) * | 2019-03-06 | 2019-06-21 | 辽宁师范大学 | Based on the multipair as altering detecting method of the extremely humorous transformation of quick quaternary number |
| CN110288592A (en) * | 2019-07-02 | 2019-09-27 | 中南大学 | A method of the zinc flotation dosing state evaluation based on probability semantic analysis model |
| CN115598025A (en) * | 2022-12-13 | 2023-01-13 | 四川亿欣新材料有限公司（Cn） | Image processing method and calcium carbonate powder quality inspection system using same |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104240297A (en) * | 2014-09-02 | 2014-12-24 | 东南大学 | Rescue robot three-dimensional environment map real-time construction method |
| CN104504715A (en) * | 2014-12-31 | 2015-04-08 | 辽宁师范大学 | Image segmentation method based on local quaternion-moment characteristic |
| CN104504698A (en) * | 2014-12-20 | 2015-04-08 | 辽宁师范大学 | Color image segmentation method based on local pixel classification |
| CN105139428A (en) * | 2015-08-11 | 2015-12-09 | 鲁东大学 | Quaternion based speeded up robust features (SURF) description method and system for color image |
| CN105608701A (en) * | 2015-12-28 | 2016-05-25 | 辽宁师范大学 | Color image segmentation method based on quaternion circular harmonic moment |
| CN105631871A (en) * | 2015-12-28 | 2016-06-01 | 辽宁师范大学 | Color image duplicating and tampering detection method based on quaternion exponent moments |
-
2016
- 2016-12-24 CN CN201611210320.5A patent/CN106815852B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104240297A (en) * | 2014-09-02 | 2014-12-24 | 东南大学 | Rescue robot three-dimensional environment map real-time construction method |
| CN104504698A (en) * | 2014-12-20 | 2015-04-08 | 辽宁师范大学 | Color image segmentation method based on local pixel classification |
| CN104504715A (en) * | 2014-12-31 | 2015-04-08 | 辽宁师范大学 | Image segmentation method based on local quaternion-moment characteristic |
| CN105139428A (en) * | 2015-08-11 | 2015-12-09 | 鲁东大学 | Quaternion based speeded up robust features (SURF) description method and system for color image |
| CN105608701A (en) * | 2015-12-28 | 2016-05-25 | 辽宁师范大学 | Color image segmentation method based on quaternion circular harmonic moment |
| CN105631871A (en) * | 2015-12-28 | 2016-06-01 | 辽宁师范大学 | Color image duplicating and tampering detection method based on quaternion exponent moments |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109919927A (en) * | 2019-03-06 | 2019-06-21 | 辽宁师范大学 | Based on the multipair as altering detecting method of the extremely humorous transformation of quick quaternary number |
| CN109919927B (en) * | 2019-03-06 | 2022-12-16 | 辽宁师范大学 | Multi-object tampering detection method based on rapid quaternion polar harmonic transformation |
| CN110288592A (en) * | 2019-07-02 | 2019-09-27 | 中南大学 | A method of the zinc flotation dosing state evaluation based on probability semantic analysis model |
| CN110288592B (en) * | 2019-07-02 | 2021-03-02 | 中南大学 | Zinc flotation dosing state evaluation method based on probability semantic analysis model |
| CN115598025A (en) * | 2022-12-13 | 2023-01-13 | 四川亿欣新材料有限公司（Cn） | Image processing method and calcium carbonate powder quality inspection system using same |
| CN115598025B (en) * | 2022-12-13 | 2023-03-10 | 四川亿欣新材料有限公司 | Image processing method and calcium carbonate powder quality inspection system using same |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106815852B (en) | 2019-08-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | Deep matching and validation network: An end-to-end solution to constrained image splicing localization and detection | |
| WO2020107717A1 (en) | Visual saliency region detection method and apparatus | |
| Wang et al. | Detection and localization of image forgeries using improved mask regional convolutional neural network | |
| CN108121991B (en) | Deep learning ship target detection method based on edge candidate region extraction | |
| CN104063706B (en) | Video fingerprint extraction method based on SURF algorithm | |
| CN103136766B (en) | A kind of object conspicuousness detection method based on color contrast and color distribution | |
| Abidin et al. | Copy-move image forgery detection using deep learning methods: a review | |
| CN105427350B (en) | Color reproduction image altering detecting method based on local quaternary number Uniformly bounded | |
| CN112528969B (en) | Face image authenticity detection method and system, computer equipment and storage medium | |
| Wang et al. | Image splicing detection based on convolutional neural network with weight combination strategy | |
| CN107066969A (en) | A kind of face identification method | |
| CN109448009A (en) | Infrared Image Processing Method and device for transmission line faultlocating | |
| CN110782442B (en) | Image artificial fuzzy detection method based on multi-domain coupling | |
| CN105139344B (en) | The method and system influenced based on frequency domain and the single image of phase equalization removal raindrop | |
| CN106600613B (en) | Improvement LBP infrared target detection method based on embedded gpu | |
| CN106815852B (en) | Color image evidence collecting method based on the extremely humorous transformation of quaternary number | |
| CN106060568A (en) | Video tampering detecting and positioning method | |
| CN104021567B (en) | Based on the fuzzy altering detecting method of image Gauss of first numeral law | |
| Li et al. | Optimized automatic seeded region growing algorithm with application to ROI extraction | |
| CN107871315B (en) | Video image motion detection method and device | |
| Zhu et al. | Shadow removal with background difference method based on shadow position and edges attributes | |
| Zhang et al. | Residential area extraction based on saliency analysis for high spatial resolution remote sensing images | |
| CN110633691A (en) | Binocular in-vivo detection method based on visible light and near-infrared camera | |
| Pal et al. | Visibility enhancement techniques for fog degraded images: a comparative analysis with performance evaluation | |
| CN114049503A (en) | Saliency region detection method based on non-end-to-end deep learning network |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190827 Termination date: 20191224 |