CN109859227A - Reproduction image detecting method, device, computer equipment and storage medium - Google Patents
Reproduction image detecting method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109859227A CN109859227A CN201910043484.0A CN201910043484A CN109859227A CN 109859227 A CN109859227 A CN 109859227A CN 201910043484 A CN201910043484 A CN 201910043484A CN 109859227 A CN109859227 A CN 109859227A
- Authority
- CN
- China
- Prior art keywords
- image
- reproduction
- straight line
- edge
- obtains
- 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of reproduction image detecting method, device, computer equipment and storage mediums, which comprises obtains test image；Edge detection is carried out to the test image using edge detection algorithm, obtains edge image；Lines detection is carried out to the edge image by line detection algorithm, obtains initial straight；Reproduction straight line is obtained from the initial straight using straight line intensive algorithms；The ratio value for calculating the quantity of the reproduction straight line and the quantity of the initial straight, as straight line ratio value, if the straight line ratio value is more than or equal to default straight line ratio value threshold value, it is determined that the test image is doubtful reproduction image.Reproduction image detection efficiency is improved using the above method, and image reproduction detection is carried out to images to be recognized by using reproduction image detecting method, improves image reproduction Detection accuracy.
Description
Technical field
The present invention relates to field of image recognition more particularly to a kind of reproduction image detecting method, device, computer equipment and
Storage medium.
Background technique
With the continuous promotion of image display technology, people can obtain the reproduction image of high quality by many methods, more
Start this kind of image being used for back door come more criminals, and current image forensics system is often difficult to detect by
Whether image belongs to reproduction.In view of reproduction image for civil order and the possible potential hazard of public safety, reproduction figure
As the research of detection becomes important project.For example, digital picture is widely used upper business bidding online now, such as utilize
Identity card image is opened an account online on line, and illegal user can be opened by the evident information found on reproduction network
Family.The image uploaded therefore, it is necessary to a kind of recognition methods of accurate reproduction image to user identifies, illegal for taking precautions against
User submits the image document forged using screen reproduction.
Traditionally, the reproduction detection and identification of image are calculated using feature extraction and neural network classification completes, framework mistake
Weight, training data are lower compared with slow and Detection accuracy, cause image reproduction detection efficiency low, influence the standard of the reproduction identification of image
True property.
Summary of the invention
The embodiment of the present invention provides a kind of reproduction image detecting method, device, computer equipment and storage medium, to solve
The not high problem of image reproduction detection efficiency.
It is situated between in addition, the embodiment of the present invention also provides a kind of reproduction image detecting method, device, computer equipment and storage
Matter, to solve the problems, such as that image reproduction detection identification accuracy is not high.
A kind of reproduction image detecting method, comprising:
Obtain test image；
Edge detection is carried out to the test image using edge detection algorithm, obtains edge image；
Lines detection is carried out to the edge image by line detection algorithm, obtains initial straight；
Reproduction straight line is obtained from the initial straight using straight line intensive algorithms；
The ratio value for calculating the quantity of the reproduction straight line and the quantity of the initial straight, as straight line ratio value, if
The straight line ratio value is more than or equal to default straight line ratio value threshold value, it is determined that the test image is doubtful reproduction figure
Picture.
A kind of reproduction image detection device, comprising:
Test image obtains module, for obtaining test image；
Edge image obtains module, for carrying out edge detection to the test image using edge detection algorithm, obtains
Edge image；
Initial straight obtains module, for carrying out lines detection to the edge image by line detection algorithm, obtains
Initial straight；
Reproduction straight line obtains module, for obtaining reproduction straight line from the initial straight using straight line intensive algorithms；
Doubtful reproduction image collection module, for calculating the quantity of the reproduction straight line and the quantity of the initial straight
Ratio value, as straight line ratio value, if the straight line ratio value is more than or equal to default straight line ratio value threshold value, it is determined that institute
Stating test image is doubtful reproduction image.
A kind of reproduction image detecting method, comprising:
Images to be recognized is obtained, and the images to be recognized is divided into N number of subgraph according to preset partitioned mode；
Each subgraph is detected using reproduction image detecting method, obtains the number of the doubtful reproduction image
Measure M, wherein M is positive integer；
The ratio value for calculating the quantity M of the doubtful reproduction image and the quantity N of subgraph, as subgraph ratio value,
If the subgraph ratio value is more than or equal to default subgraph ratio value threshold value, the images to be recognized is reproduction figure
Picture.
A kind of reproduction image detection device, comprising:
Subgraph obtains module, for obtaining images to be recognized, and by the images to be recognized according to preset piecemeal side
Formula is divided into N number of subgraph；
The quantity of doubtful reproduction image obtains module, for using each subgraph such as any one of right 1-4 institute
The reproduction image detecting method stated is detected, and obtains the quantity M of the doubtful reproduction image, wherein M is positive integer；
Reproduction picture recognition module, for calculating the ratio of the quantity M of the doubtful reproduction image and the quantity N of subgraph
Value, it is described if the subgraph ratio value is more than or equal to default subgraph ratio value threshold value as subgraph ratio value
Images to be recognized is reproduction image.
A kind of computer equipment, including memory, processor and storage are in the memory and can be in the processing
The computer program run on device, the processor realize above-mentioned reproduction image detecting method when executing the computer program,
Alternatively, the processor realizes above-mentioned reproduction image detecting method when executing the computer program.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter
Calculation machine program realizes above-mentioned reproduction image detecting method when being executed by processor, alternatively, the processor executes the computer
Above-mentioned reproduction image detecting method is realized when program.
In above-mentioned reproduction image detecting method, device, computer equipment and storage medium, test image is obtained first, so
Edge detection is carried out to test image using edge detection algorithm afterwards, obtains edge image, is based on the edge image so as to subsequent
Obtain more accurate reproduction feature；Then lines detection is carried out to edge image by line detection algorithm, obtained initial straight
Line extracts more accurate reproduction pixel characteristic in initial straight so as to subsequent；Next, using straight line intensive algorithms from initial straight
Reproduction straight line is obtained in line, is conducive to raising reproduction image detection and is obtained accuracy；Finally calculate the quantity of reproduction straight line and initial
The ratio value of the quantity of straight line is more than or equal to default straight line ratio value threshold value in straight line ratio value as straight line ratio value
When, determine that test image is doubtful reproduction image, not only algorithm is simple, but also must improve the effect of reproduction image detection significantly
Rate.
In above-mentioned reproduction image detecting method, device, computer equipment and storage medium, images to be recognized is obtained first,
And images to be recognized is divided into N number of subgraph according to preset partitioned mode, subimage block contains richer part letter
Breath, so that the accuracy rate of subsequent reproduction image detection is improved.Then reproduction image detecting method is used to each subgraph
It is detected, obtains the quantity M of doubtful reproduction image.Finally calculate the quantity M's of the doubtful reproduction image and quantity N of subgraph
Ratio value, as subgraph ratio value, if subgraph ratio value is more than or equal to default subgraph ratio value threshold value, wait know
Other image is reproduction image.Since reproduction image detecting method accuracy rate is high, so that the accuracy rate of doubtful reproduction image detection is high,
And it fully considers the accurate local message of more horn of plenty in subgraph, has improved the accuracy of image reproduction image detection.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the application environment schematic diagram of reproduction image detecting method provided in an embodiment of the present invention；
Fig. 2 is one exemplary diagram of reproduction image detecting method provided in an embodiment of the present invention；
Fig. 3 is another exemplary diagram of reproduction image detecting method provided in an embodiment of the present invention；
Fig. 4 is another exemplary diagram of reproduction image detecting method provided in an embodiment of the present invention；
Fig. 5 is another exemplary diagram of reproduction image detecting method provided in an embodiment of the present invention；
Fig. 6 is a functional block diagram of reproduction image detection device provided in an embodiment of the present invention；
Fig. 7 is another functional block diagram of reproduction image detection device provided in an embodiment of the present invention；
Fig. 8 is one exemplary diagram of reproduction image detecting method provided in an embodiment of the present invention；
Fig. 9 is another functional block diagram of reproduction image detection device provided in an embodiment of the present invention；
Figure 10 is a schematic diagram of computer equipment provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Reproduction image detecting method provided by the present application, can be applicable in the application environment such as Fig. 1, wherein client is logical
It crosses network to be communicated with server-side, server-side receives the test image that client is sent, using edge detection algorithm to test
Image carries out edge detection, obtains edge image.Then lines detection is carried out to edge image by line detection algorithm, obtained
Initial straight.Then reproduction straight line is obtained from initial straight using straight line intensive algorithms.And then calculate the quantity of reproduction straight line
It is more than or equal to default straight line ratio in straight line ratio value as straight line ratio value with the ratio value of the quantity of initial straight
When being worth threshold value, determine that test image is doubtful reproduction image.Wherein, client can be, but not limited to be various personal computers,
Laptop, smart phone, tablet computer and portable wearable device.Server-side can be with independent server either
The server cluster of multiple servers composition is realized.
In one embodiment, as shown in Fig. 2, being applied to be illustrated for the server-side in Fig. 1 in this way, including
Following steps:
S10: test image is obtained.
Wherein, test image is the two dimensional image acquired using Image Acquisition tool, such as outdoor scene, indoor scene and people
The two dimensional images such as object, for as the image for carrying out reproduction image detection.Optionally, test image can be through client reality
When the image that acquires, be also possible to be stored in advance image on the client.Specifically, test image is sent to clothes by client
Business end, server-side get the test image.
S20: edge detection is carried out to test image using edge detection algorithm, obtains edge image.
Wherein, the edge of image refers to the significant part of image regional area brightness change, the gray scale section one in the region
As can be regarded as a step, i.e., from a gray value in the buffer area of very little change dramatically to another gray scale differ
Biggish gray value.Edge image refers to the edge contour image of test image.It is to be appreciated that the marginal portion of Image boundaries
Most of profile information of the test chart as itself, identification of the determination and extraction of Image boundaries for entire image scene are concentrated
It is very important with understanding, while being also the important feature that Image is relied on.Edge detection is to test chart as carrying out
Measurement, detection and the positioning of grey scale change.Edge detection algorithm can be Sobel edge detection algorithm, be also possible to the side Canny
Edge detection algorithm can also be Roberts edge detection algorithm etc..
It is worth noting that the edge detection in this step is the reproduction pixel characteristic for obtaining image, and reproduction picture
The acquisition of plain feature, key be to get the real edge image of test image, thus, the embodiment of the present invention provides a kind of excellent
Scheme is selected, i.e., obtains the edge image of test image using Canny edge detection algorithm.It is to be appreciated that Canny edge detection
Algorithm has used two different threshold values to detect strong edge and weak edge respectively, and only when weak edge is connected with strong edge,
It include to export in image again by weak edge, therefore, it is more to have that Canny edge detection algorithm, which is able to detect that real weak edge,
The edge detection algorithm of effect, must can improve the efficiency of edge detection algorithm significantly, while test image can occur due to reproduction
The phenomenon that fogging, and this phenomenon is in the more obvious of the edge of image embodiment.In the present embodiment, by using edge
Detection algorithm carries out edge detection to test image, obtains edge image, so as to subsequent more quasi- based on edge image acquisition
True reproduction pixel characteristic.
S30: lines detection is carried out to edge image by line detection algorithm, obtains initial straight.
Wherein, line detection algorithm refers to that the detection method for extracting straight line present in image, line detection algorithm include
But it is not limited to Hough (Hough transformation) line detection algorithm, Freeman (chain code) line detection algorithm or looper crawling to calculate
Method.Initial straight refers to the straight line that the pixel for meeting straight line condition in edge image connects, to extract test image
Reproduction pixel characteristic provides the foundation.Specifically, it is extracted by line detection algorithm and is met present in connection edge image
The straight line that the pixel of straight line condition is formed, as initial straight.The pixel therein for meeting straight line condition namely satisfaction are every
The equal pixel of slope between two pixels.
It is to be appreciated that since reproduction image is to be scanned, print or other equipment with shooting function are to true
Real image carries out reproduction, it is contemplated that during carrying out reproduction to true picture, shows the characteristic and reproduction of medium itself
The scene of journey is distinguished, so that reproduction image has differences with true picture.Such as: reproduction anamorphose, reproduction imaging surface ladder
Angle value can generate nonlinear change compared with true picture, and it is abnormal that this generates reproduction imaging surface gradient value, and then causes to turn over
It claps the distribution of initial straight present in image to change, therefore, the initial straight in edge image is extracted, so as to subsequent first
The more accurate reproduction pixel characteristic of beginning lines detection.
S40: reproduction straight line is obtained from initial straight using straight line intensive algorithms.
Wherein, reproduction straight line, which refers to, carries out lines detection using straight line intensive algorithms to initial straight, and having for obtaining is turned over
The straight line for clapping pixel characteristic, for as the reproduction pixel characteristic for distinguishing reproduction image and non-reproduction image (true picture).Tool
Body, the reproduction straight line with reproduction pixel characteristic is extracted using straight line intensive algorithms in initial straight.
Wherein, straight line intensive algorithms are a kind of algorithms for extracting and meeting the straight line of straight line dense convergence and parallel rule, are used
Reproduction pixel characteristic in extraction straight line.Reproduction straight line refers to the straight line for meeting straight line intensive algorithms criterion, i.e. straight line
Dense convergence and the parallel criterion include less than 1 ° of the slope difference (degree) of two straight lines, and adjacent parallel two are straight
The distance of line is less than preset distance threshold.In straight line intensive algorithms, initial straight needs to meet " parallel " and " dense convergence "
The two conditions.For " parallel " this condition, i.e., after initial straight pixel is written in the slope value of two initial straights, if
Less than 1 ° of slope difference (degree), then initial straight is parallel, that is, meets the condition of " parallel ".For " dense convergence " this condition,
The distance between two parallel lines (initial straight) are calculated, the distance will be met directly less than two of preset distance threshold
Line is determined as meeting " dense convergence " this condition namely reproduction straight line.If the linear equation of initial straight L1 is k1x-d1=y,
The linear equation of initial straight L2 is k2x-d2=y, and initial straight L1 is adjacent with the position initial straight L2, the public affairs of criterion
Formula is as follows:
Wherein, k1For the slope of initial straight L1, k2For the slope of initial straight L2, d1For the intercept of initial straight L1, d2
For the intercept of initial straight L2, α is preset distance threshold.
Specifically, the slope for calculating initial straight, by the criterions of straight line intensive algorithms to initial straight slope into
Row differentiates, that is, judges whether any two adjacent initial straights meet above-mentioned criterion formula, if satisfied, being determined as reproduction
Straight line.The initial straight of the criterion of straight line intensive algorithms will be met as reproduction straight line, it is possible to understand that ground, due to really scheming
Therefore the difference of picture and reproduction image on pixel block feature when initial straight meets straight line dense convergence and parallel, then should
Straight line is determined as reproduction straight line, so that the confirmation of reproduction pixel characteristic is more accurate.And using straight line intensive algorithms from initial
Reproduction straight line is obtained in straight line, is conducive to the accuracy for improving reproduction image detection.
S50: the ratio value of the quantity of reproduction straight line and the quantity of initial straight is calculated, as straight line ratio value, if straight line
Ratio value is more than or equal to default straight line ratio value threshold value, it is determined that test image is doubtful reproduction image.
Wherein, straight line ratio value refers to the ratio value of the quantity of reproduction straight line and the quantity of initial straight.For example, reproduction is straight
The quantity of line is a, and the quantity of initial straight is b, then straight line ratio value isSince reproduction straight line is extracted in initial straight
, therefore, a≤b, the i.e. value range of straight line ratio value are [0,1].It specifically, can be by accumulator to reproduction straight line
The quantity of quantity and initial straight is calculated, and the quantity of reproduction straight line and the quantity of initial straight are obtained.Default straight line ratio
Value threshold value is to detect the critical value of the straight line ratio value of reproduction image and true picture, for determining that test image is reproduction image
Or the boundary of true picture.Doubtful reproduction image refers to that reproduction pixel characteristic quantity reaches the image of preset condition, i.e. straight line
Ratio value reaches the image of threshold value.For example, the quantity of the initial straight of a test image is 100, the quantity of reproduction straight line is 20,
Default straight line ratio value threshold value is 10%, and the ratio value of the quantity of the quantity and initial straight of reproduction straight line is 20%, is greater than pre-
If straight line ratio value threshold value 10%, therefore, which is doubtful reproduction image.
It is worth noting that the image of different resolution, default straight line ratio value threshold value is different, can be according to reality
Application demand setting, herein with no restriction.
In step S50, by calculating the ratio value of the quantity of reproduction straight line and the quantity of initial straight, as straight line ratio
Value, and the test image that straight line ratio value is more than or equal to default straight line ratio value threshold value is confirmed as doubtful reproduction image.
According to the quantity and proportionate relationship of reproduction pixel characteristic, to detect test image, not only algorithm is simple, but also must improve significantly
Reproduction image detection obtains efficiency.
In the present embodiment, test image is obtained first, and edge inspection is then carried out to test image using edge detection algorithm
It surveys, obtains edge image, more accurate reproduction feature is obtained based on the edge image so as to subsequent；Then pass through straight-line detection
Algorithm carries out lines detection to edge image, obtains initial straight, extracts more accurate reproduction picture in initial straight so as to subsequent
Plain feature；Next, obtaining reproduction straight line from initial straight using straight line intensive algorithms, be conducive to improve reproduction image detection
Obtain accuracy；The ratio value for finally calculating the quantity of reproduction straight line and the quantity of initial straight, as straight line ratio value, in straight line
When ratio value is more than or equal to default straight line ratio value threshold value, determine that test image is doubtful reproduction image, not only algorithm letter
It is single, and the efficiency of reproduction image detection must be improved significantly.
In one embodiment, as shown in figure 3, in step S20, edge inspection is carried out to test image using edge detection algorithm
It surveys, obtains edge image, comprising:
S21: test image is smoothed using Gaussian filter, obtains smoothed image.
Wherein, Gaussian filter is a kind of for eliminating the filter of image Gaussian noise.Smoothing processing refers to test
Image is removed the operation of noise, for reducing interference of the noise to test image.It specifically, can be by Python
Image mean value smoothing filtering GaussianBlur () function is smoothed, so that the information of smoothed image more horn of plenty, example
Such as:
Blur=cv2.GaussianBlur (img, (5,5), 0)；
Wherein, img is test image title, and Blur is smoothed image title, and (5,5) refer to that window size is for 5 × 5,0
Refer to the standard deviation of Gaussian function.It is to be appreciated that above-mentioned numerical value is only an illustrative explanation, it specifically can be according to practical feelings
Condition is configured and adjusts.
S22: gradient magnitude and direction using single order local derviation finite difference formulations smoothed image obtain gradient information.
Wherein, gradient information refers to gradient magnitude and the direction of smoothed image.It is flat using single order local derviation finite difference formulations
The gradient magnitude of sliding image and direction include following calculating process:
The gradient of each pixel is obtained first, if the pixel of smoothed image is (x, y), gradient calculation formula is as follows:
Wherein, Gx(x, y) is the horizontal direction gradient magnitude of the pixel (x, y) of smoothed image, wherein Gy(x, y) is smooth
The vertical gradient amplitude of the pixel (x, y) of image, H (x, y) are the gray value of pixel (x, y).
Then it is calculated using the following equation the pixel gradient size of the smoothed image:
Wherein, G (x, y) is the gradient magnitude of smoothed image.
Finally, being calculated using the following equation the gradient direction of smoothed image:
Wherein, α (x, y) is the deflection of the gradient direction of smoothed image.
It is to be appreciated that gray-scale intensity changes strongest position in the corresponding smoothed image of gradient direction α (x, y).So-called change
Change is most strong, that is, refers to gradient direction.Single order local derviation finite difference formulations horizontal direction and vertical direction are used to smoothed out image
First derivative (image gradient), so as to subsequent gradient and the direction for determining boundary according to gradient information.
S23: non-maxima suppression is carried out to gradient information, obtains K marginal point, K is positive integer.
Wherein, non-maxima suppression is that the boundary for smoothed image to be obscured (blurred) is apparent from (sharp).
Retain the maximum of gradient intensity on each pixel, and deletes other values.Specifically, right after obtaining gradient information
Gradient information carries out non-maxima suppression, obtains the maximum pixel of pixel value in K identical gradient directions, i.e. K side
Edge point, wherein K refers to the quantity of marginal point, and the numerical value of K is determined by gradient information, herein with no restrictions.
S24: the strong edge point in K marginal point is connected with dual threashold value-based algorithm, obtains edge image.
Wherein, dual threashold value-based algorithm refers to the algorithm that bound pair pixel is chosen under one threshold value upper bound of setting and threshold value.
Specifically, two threshold values: maxVal (the threshold value upper bound) and minVal (threshold value lower bound) are set.By each marginal point respectively and
MaxVal (the threshold value upper bound) and minVal (threshold value lower bound) are compared, i.e., when the shade of gray of marginal point is greater than maxVal,
Confirm that the marginal point is strong edge point, when the shade of gray of marginal point is less than minVal, confirm that the marginal point is weak marginal point,
And delete the weak marginal point.When the shade of gray of marginal point is being less than or equal to maxVal and is being more than or equal to minVal
When, then judge whether the marginal point is connected with strong edge point, if the marginal point is connected with strong edge point, confirms that the marginal point is
Otherwise strong edge point confirms that the marginal point is weak marginal point, and delete the weak marginal point.And then strong edge point is filtered out, and will
Each strong edge point is attached, and forms edge image.It therefore, can be by strong edge point and weak marginal point by dual threashold value-based algorithm
It distinguishes, so that the edge image detected is more accurate.
In the present embodiment, test image is smoothed using Gaussian filter first, obtains smoothed image, is reduced
Interference of the noise to test image, so that the information of smoothed image more horn of plenty；Then single order local derviation finite difference formulations are used
The gradient magnitude of smoothed image and direction, obtain gradient information, so as to subsequent gradient and the side for determining boundary according to gradient information
To；Then non-maxima suppression is carried out to gradient information, obtains K marginal point；Finally, connecting K edge with dual threashold value-based algorithm
Strong edge point in point, obtains edge image, can distinguish strong edge point and strong edge point by dual threashold value-based algorithm, make
The edge image that must be detected is more accurate.
In one embodiment, as shown in figure 4, in step S30, straight line is carried out to edge image by line detection algorithm and is mentioned
It takes, obtains initial straight, specifically comprise the following steps:
S31: edge image is converted into the image of bi-level fashion, obtains binary edge map.
Wherein, the image of bi-level fashion refers to that the gray value of each pixel is black or white, and gray value is without in
Between transition image, for descriptive text or the profile of image.Binary edge map refers to the edge image of bi-level fashion.Tool
Body, the profile of edge image is described by the image of bi-level fashion, to survey subsequently through the contours extract of test image
Attempt the reproduction pixel characteristic of picture.It is to be appreciated that the profile of image includes the relatively multi information of image, therefore, by by side
Edge image is converted into the image of bi-level fashion, is conducive to extract more accurate reproduction pixel characteristic.
S32: the pixel data of binary edge map is obtained.
Wherein, pixel data refers to the data information of the pixel of binary edge map, such as the brightness value of pixel.Specifically
Ground, obtaining for the pixel data of binary edge map can be obtained tool acquisition by the pixel in the library Opencv, can also used
Pixel in matlab obtains tool and obtains, and can be selected according to actual needs, herein with no restrictions.
In the present embodiment, by obtaining the pixel data of binary edge map, turned over so as to subsequent based on pixel data extraction
Clap pixel characteristic.
S33: Hough transformation is carried out to pixel data according to Hough transformation formula, obtains initial straight.
Wherein, Hough transformation (Hough Transform) is one of image procossing Feature Extraction Technology, it passes through one
Kind Voting Algorithm detection has the object of specific shape.Wherein, the Hough transformation formula of straight line refers to through Hough transformation principle
To detect the discrimination formula that whether there is straight line in edge image.Hough transformation formula is as shown in following formula:
Xcos θ+ysin θ=ρ ρ >=0,0≤θ≤π；
Wherein, ρ is distance of the straight line L to origin；θ is the angle of straight line L and positive direction of the x-axis, according to above-mentioned formula, straight line
The upper different point of L is transformed to the sine curve that cluster intersects at (x, y) point in parameter plane ρ-θ.According in image space
Data point calculation Hough parameter space in sinusoidal path, carry out Two-dimensional Statistical in parameter plane, choose peak value.
The peak value is exactly the parameter of straight line in image space, specifically, using pixel data as the input of Hough transformation formula, is led to
Cross Hough transformation, obtain initial straight, to realize the straight-line detection in image space, straight line quantity that detection is obtained into
Row is cumulative, and then determines the quantity of the initial straight of edge image.
Specifically, HoughLines () function and HoughLinesP () function may be implemented respectively in opencv
Hough straight-line detection.Wherein, HoughLines () function implementation is standard Hough transformation.Standard Hough transformation be by
Edge image is mapped on its parameter space, by calculating all marginal points, determines initial straight.It can be rapidly and accurately
Determine initial straight.
In the present embodiment, edge image is converted into the image of bi-level fashion first, obtains binary edge map, is conducive to
Extract more accurate reproduction pixel characteristic.Then, by obtaining the pixel data of binary edge map, to extract reproduction picture
Plain feature.Hough transformation is finally carried out to pixel data according to Hough transformation formula, obtains initial straight, it can be rapidly and accurately
Obtain initial straight.
In one embodiment, as shown in figure 5, in step S30, straight line is carried out to edge image by line detection algorithm and is mentioned
It takes, obtains initial straight, specifically comprise the following steps:
S31 ': Chain Code Detection is carried out to edge image, obtains the chain code of edge image, and calculate the histogram of chain code.
Wherein, chain code is also known as freeman code, is to describe song with the coordinate of curve starting point and edge direction code
The method on line or boundary, for indicating the coded representation of curve and zone boundary.Chain Code Detection, which refers to, carries out chain code to image
After tracking is connected with chain code, detection obtains the processing of the chain code of image,.Wherein, the histogram for calculating chain code can be according to histogram
The definition of figure is calculated, such as: to the chain code { a of edge imagei}n(ai∈ { 0,1,2,3,4,5,6,7 }) each of member
Element takes K point before the point and K point after the point calculates the histogram of chain code forward at the point and chain code is straight backward
Fang Tu obtains chain code histogram.
Specifically, by detecting the boundary line of edge image, the chain code of edge image is obtained, and calculate the histogram of chain code
Figure obtains straight line parameter based on chain code histogram so as to subsequent.
S32 ': the histogram based on chain code obtains the straight line parameter of straight line chain code.
Wherein, straight line parameter refers to the straight line parameter on two-dimensional surface, including this 4 ginsengs of starting point coordinate and terminal point coordinate
Number.Specifically, the histogram of chain code forward of the chain code in calculating step S31 ' and the backward maximum of chain code histogram, when forward
When the maximum of chain code histogram and backward chain code histogram is equal, illustrate that the chain code is straight line, and then it is straight to extract straight line place
The straight line parameter of line segment, i.e. starting point coordinate and terminal point coordinate.
S33 ': the straight line that straight line parameter is determined is as initial straight.
Specifically, obtain one group of linear equation, the linear equation is true is estimated using least square method to straight line parameter
Fixed straight line is initial straight.By estimating straight line parameter, deviation caused by determining straight line is effectively overcome, and make
Straight line parameter information is all efficiently used, and the fitting precision of initial straight is improved.
In the present embodiment, Chain Code Detection is carried out to edge image first, obtains the chain code of edge image, and calculate chain code
Histogram.It is then based on the histogram of chain code, obtains the straight line parameter of straight line chain code.The straight line that finally straight line parameter is determined is made
For initial straight, deviation caused by determining straight line is effectively overcome, and straight line parameter information is all efficiently used, improve
The fitting precision of initial straight.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
In one embodiment, a kind of reproduction image detection device is provided, the reproduction image detection device and above-described embodiment
Middle reproduction image detecting method corresponds.As shown in fig. 6, the reproduction image detection device includes that test image obtains module
10, edge image obtains module 20, initial straight obtains module 30, reproduction straight line obtains module 40 and doubtful reproduction image obtains
Module 50.Detailed description are as follows for each functional module:
Test image obtains module 10, for obtaining test image；
Edge image obtains module 20, for carrying out edge detection to test image using edge detection algorithm, obtains side
Edge image；
Initial straight obtains module 30, for carrying out lines detection to edge image by line detection algorithm, obtains just
Beginning straight line；
Reproduction straight line obtains module 40, for obtaining reproduction straight line from initial straight using straight line intensive algorithms；
Doubtful reproduction image collection module 50, for calculating the ratio of the quantity of reproduction straight line and the quantity of initial straight
Value, as straight line ratio value, if straight line ratio value is more than or equal to default straight line ratio value threshold value, it is determined that test image is
Doubtful reproduction image.
Preferably, it includes that smoothed image acquiring unit, gradient information acquiring unit, marginal point obtain that edge image, which obtains module,
Take unit and edge image acquiring unit.
Smoothed image acquiring unit is smoothly schemed for being smoothed using Gaussian filter to test image
Picture；
Gradient information acquiring unit, for using gradient magnitude and the side of single order local derviation finite difference formulations smoothed image
To obtaining gradient information；
Marginal point acquiring unit obtains K marginal point, K is positive whole for carrying out non-maxima suppression to gradient information
Number；
Edge image acquiring unit obtains edge graph for connecting the strong edge point in K marginal point with dual threashold value-based algorithm
Picture.
Preferably, initial straight obtain module include binary edge map acquiring unit, pixel data acquiring unit and just
Beginning straight line acquiring unit.
Binary edge map acquiring unit obtains binary edge for edge image to be converted into the image of bi-level fashion
Image；
Pixel data acquiring unit, for obtaining the pixel data of binary edge map；
Initial straight acquiring unit obtains initial for carrying out Hough transformation to pixel data according to Hough transformation formula
Straight line.
Preferably, as shown in fig. 7, initial straight obtains histogram calculation unit 31 ', the straight line that module 30 further includes chain code
Parameter acquiring unit 32 ' and initial straight acquiring unit 33 '.
The histogram calculation unit 31 ' of chain code obtains the chain of edge image for carrying out Chain Code Detection to edge image
Code, and calculate the histogram of chain code；
Straight line parameter acquiring unit 32 ' obtains the straight line parameter of straight line chain code for the histogram based on chain code；
Initial straight acquiring unit 33 ', the straight line for determining straight line parameter is as initial straight.
In one embodiment, a reproduction image detecting method is provided, which can also apply such as
In the application environment of Fig. 1, wherein client is communicated by network with server-side.Server-side receive client send to
It identifies image, and images to be recognized is divided into N number of subgraph according to preset partitioned mode.Then it obtains and each subgraph is adopted
It is detected with reproduction image detecting method, obtains the quantity M of doubtful reproduction image, finally calculate the quantity of doubtful reproduction image
The ratio value of the quantity N of M and subgraph, as subgraph ratio value, if subgraph ratio value is more than or equal to default subgraph
As ratio value threshold value, then images to be recognized is reproduction image.Wherein, client can be, but not limited to be various personal computers,
Laptop, smart phone, tablet computer and portable wearable device.Server-side can be with independent server either
The server cluster of multiple servers composition is realized.
In one embodiment, as shown in figure 8, being applied to be illustrated for the server-side in Fig. 1 in this way, including
Following steps:
S60: images to be recognized is obtained, and images to be recognized is divided into N number of subgraph according to preset partitioned mode.
Wherein, images to be recognized refer to carry out identification image whether be reproduction image image.Subgraph is will be to
Identify each block of image after image divides.Preset partitioned mode refers to preset partitioned mode, such as by figure to be identified
As carrying out piecemeal according to the partitioned mode of 9 × 9 equal parts, the quantity of subgraph is 81 namely N=81.It for another example, will be to be identified
Image carries out piecemeal according to lateral X equal part, the partitioned mode of longitudinal direction Y equal part, and X is positive integer, and Y is positive integer, at this point, subgraph
Quantity be X × Y namely N=X × Y.In the present embodiment, by being drawn to images to be recognized according to preset partitioned mode
It is divided into N number of subgraph, reproduction image detection is carried out based on each subimage block so as to subsequent, it is possible to understand that ground, subimage block packet
Richer local message is contained, so that the accuracy rate of subsequent reproduction image detection is improved.
Preset partitioned mode can be set according to the resolution ratio of image to be detected, herein with no restrictions.
S70: being detected each subgraph using reproduction image detecting method, obtain the quantity M of doubtful reproduction image,
Wherein, M is positive integer.
Specifically, each subimage block is detected using reproduction image detecting method respectively, if subgraph is doubtful
Reproduction image, then the quantity of doubtful reproduction image adds 1, if subgraph is not doubtful reproduction image, the number of doubtful reproduction image
It measures constant.For example, there is 27 doubtful reproduction images during 81 subgraphs are fast, then M=27.
S80: calculating the ratio value of the quantity M of doubtful reproduction image and the quantity N of subgraph, as subgraph ratio value,
If subgraph ratio value is more than or equal to default subgraph ratio value threshold value, images to be recognized is reproduction image.
Specifically, presetting subgraph ratio value threshold value is whether preset images to be recognized for identification is reproduction figure
The critical value of the subgraph ratio value of picture.In a specific embodiment, presetting subgraph ratio value threshold value is 20%, subgraph
Ratio value is 27/81=30%, so the images to be recognized is reproduction image.In the present embodiment, by counting subgraph quantity
And doubtful reproduction amount of images judges whether the identification image is reproduction image, due to reproduction figure according to the ratio value of the two
As detection method accuracy rate height, so that the accuracy rate of doubtful reproduction image detection is high, and fully consider in subgraph more
Accurate local message is enriched, the accuracy of image reproduction image detection is improved.
In the present embodiment, images to be recognized is obtained first, and images to be recognized is divided into N according to preset partitioned mode
A subgraph, subimage block contain richer local message, so that the accuracy rate of subsequent reproduction image detection is improved.
Then each subgraph is detected using reproduction image detecting method, obtains the quantity M of doubtful reproduction image.Finally calculate
The ratio value of the quantity N of the quantity M and subgraph of doubtful reproduction image, as subgraph ratio value, if subgraph ratio value is big
In or equal to default subgraph ratio value threshold value, then images to be recognized is reproduction image.Since reproduction image detecting method is quasi-
True rate is high, so that the accuracy rate of doubtful reproduction image detection is high, and has fully considered the accurate office of more horn of plenty in subgraph
Portion's information improves the accuracy of image reproduction image detection.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
In one embodiment, a kind of reproduction image detection device is provided, the reproduction image detection device and above-described embodiment
Middle reproduction image detecting method corresponds.As shown in figure 9, the reproduction image detection device include subgraph obtain module 60,
The quantity of doubtful reproduction image obtains module 70 and reproduction picture recognition module 80.Detailed description are as follows for each functional module:
Subgraph obtains module 60, for obtaining images to be recognized, and by images to be recognized according to preset partitioned mode
It is divided into N number of subgraph；
The quantity of doubtful reproduction image obtains module 70, for being carried out to each subgraph using reproduction image detecting method
Detection, obtains the quantity M of doubtful reproduction image, wherein M is positive integer；
Reproduction picture recognition module 80, for calculating the ratio of the quantity M of doubtful reproduction image and the quantity N of subgraph
Value, as subgraph ratio value, if subgraph ratio value is more than or equal to default subgraph ratio value threshold value, figure to be identified
As being reproduction image.
Specific about reproduction detection device limits the restriction that may refer to above for reproduction image detecting method,
This is repeated no more.Modules in above-mentioned reproduction detection device can come real fully or partially through software, hardware and combinations thereof
It is existing.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also be with software shape
Formula is stored in the memory in computer equipment, executes the corresponding operation of the above modules in order to which processor calls.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 10.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is for storing test image and images to be recognized.The network interface of the computer equipment is used for and outside
Terminal passes through network connection communication.To realize a kind of reproduction image detecting method when the computer program is executed by processor.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor realize the reproduction figure in above-described embodiment when executing computer program
As detection method.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program realizes the reproduction image detecting method in above-described embodiment when being executed by processor.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Instruct relevant hardware to complete by computer program, computer program to can be stored in a non-volatile computer readable
It takes in storage medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, this Shen
Please provided by any reference used in each embodiment to memory, storage, database or other media, may each comprise
Non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of device are divided into different functional unit or module, to complete above description
All or part of function.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification or
Replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all include
Within protection scope of the present invention.
Claims (10)
1. a kind of reproduction image detecting method, which is characterized in that the reproduction image detecting method includes:
Obtain test image；
Edge detection is carried out to the test image using edge detection algorithm, obtains edge image；
Lines detection is carried out to the edge image by line detection algorithm, obtains initial straight；
Reproduction straight line is obtained from the initial straight using straight line intensive algorithms；
The ratio value for calculating the quantity of the reproduction straight line and the quantity of the initial straight, as straight line ratio value, if described
Straight line ratio value is more than or equal to default straight line ratio value threshold value, it is determined that the test image is doubtful reproduction image.
2. reproduction image detecting method as described in claim 1, which is characterized in that described to use edge detection algorithm to described
Test image carries out edge detection, obtains edge image, comprising:
The test image is smoothed using Gaussian filter, obtains smoothed image；
Gradient magnitude and direction using smoothed image described in single order local derviation finite difference formulations, obtain gradient information；
Non-maxima suppression is carried out to the gradient information, obtains K marginal point, K is positive integer；
With the strong edge point in K marginal points of dual threashold value-based algorithm connection, the edge image is obtained.
3. reproduction image detecting method as claimed in claim 2, which is characterized in that it is described by line detection algorithm to described
Edge image carries out lines detection, obtains initial straight, comprising:
The edge image is converted into the image of bi-level fashion, obtains binary edge map；
Obtain the pixel data of the binary edge map；
Hough transformation is carried out to the pixel data according to Hough transformation formula, obtains the initial straight.
4. reproduction image detecting method as described in claim 1, which is characterized in that it is described by line detection algorithm to described
Edge image carries out lines detection, obtains initial straight, comprising:
Chain Code Detection is carried out to the edge image, obtains the chain code of the edge image, and calculate the histogram of the chain code；
Based on the histogram of the chain code, the straight line parameter of straight line chain code is obtained；
The straight line that the straight line parameter is determined is as the initial straight.
5. a kind of reproduction image detecting method, which is characterized in that the reproduction recognition methods includes:
Images to be recognized is obtained, and the images to be recognized is divided into N number of subgraph according to preset partitioned mode；
Each subgraph is detected using reproduction image detecting method according to any one of claims 1-4, is obtained
Take the quantity M of the doubtful reproduction image, wherein M is positive integer；
The ratio value for calculating the quantity M of the doubtful reproduction image and the quantity N of subgraph, as subgraph ratio value, if institute
It states subgraph ratio value and is more than or equal to default subgraph ratio value threshold value, then the images to be recognized is reproduction image.
6. a kind of reproduction image detection device, which is characterized in that the reproduction image detection device includes:
Test image obtains module, for obtaining test image；
Edge image obtains module, for carrying out edge detection to the test image using edge detection algorithm, obtains edge
Image；
Initial straight obtains module, for carrying out lines detection to the edge image by line detection algorithm, obtains initial
Straight line；
Reproduction straight line obtains module, for obtaining reproduction straight line from the initial straight using straight line intensive algorithms；
Doubtful reproduction image collection module, for calculating the ratio of the quantity of the reproduction straight line and the quantity of the initial straight
Value, as straight line ratio value, if the straight line ratio value is more than or equal to default straight line ratio value threshold value, it is determined that the survey
Attempt as being doubtful reproduction image.
7. image reproduction image detection device as claimed in claim 6, which is characterized in that the initial straight obtains module,
Include:
The histogram calculation unit of chain code obtains the chain of the edge image for carrying out Chain Code Detection to the edge image
Code, and calculate the histogram of the chain code；
Straight line parameter acquiring unit obtains the straight line parameter of straight line chain code for the histogram based on the chain code；
Initial straight acquiring unit, the straight line for determining the straight line parameter is as the initial straight.
8. a kind of reproduction image detection device, which is characterized in that the reproduction identification device includes:
Subgraph obtains module, draws for obtaining images to be recognized, and by the images to be recognized according to preset partitioned mode
It is divided into N number of subgraph；
The quantity of doubtful reproduction image obtains module, for using each subgraph such as any one of claim 1-4 institute
The reproduction image detecting method stated is detected, and obtains the quantity M of the doubtful reproduction image, wherein M is positive integer；
Reproduction picture recognition module, for calculating the ratio value of the quantity M of the doubtful reproduction image and the quantity N of subgraph,
As subgraph ratio value, if the subgraph ratio value is more than or equal to default subgraph ratio value threshold value, it is described to
Identification image is reproduction image.
9. a kind of computer equipment, including memory, processor and storage are in the memory and can be in the processor
The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to
4 described in any item reproduction image detecting methods or the processor are realized when executing the computer program as right is wanted
Reproduction image detecting method described in asking 5.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In, the reproduction image detecting method as described in any one of Claims 1-4 is realized when the computer program is executed by processor,
Alternatively, the computer program realizes reproduction image detecting method as claimed in claim 5 when being executed by processor.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910043484.0A CN109859227B (en) | 2019-01-17 | 2019-01-17 | Method and device for detecting flip image, computer equipment and storage medium |
| PCT/CN2019/117396 WO2020147398A1 (en) | 2019-01-17 | 2019-11-12 | Reproduced image detection method and device, computer device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910043484.0A CN109859227B (en) | 2019-01-17 | 2019-01-17 | Method and device for detecting flip image, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109859227A true CN109859227A (en) | 2019-06-07 |
| CN109859227B CN109859227B (en) | 2023-07-14 |
Family
ID=66894943
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910043484.0A Active CN109859227B (en) | 2019-01-17 | 2019-01-17 | Method and device for detecting flip image, computer equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109859227B (en) |
| WO (1) | WO2020147398A1 (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110348511A (en) * | 2019-07-08 | 2019-10-18 | 创新奇智(青岛)科技有限公司 | A kind of picture reproduction detection method, system and electronic equipment |
| CN110706277A (en) * | 2019-09-12 | 2020-01-17 | 创新奇智(合肥)科技有限公司 | Method for detecting object rotation angle based on chain code histogram |
| WO2020147398A1 (en) * | 2019-01-17 | 2020-07-23 | 平安科技（深圳）有限公司 | Reproduced image detection method and device, computer device and storage medium |
| CN111461143A (en) * | 2020-03-31 | 2020-07-28 | 珠海格力电器股份有限公司 | Picture copying identification method and device and electronic equipment |
| CN112115945A (en) * | 2020-09-25 | 2020-12-22 | 有米科技股份有限公司 | Method and device for extracting dynamic and static subtitles |
| CN113542142A (en) * | 2020-04-14 | 2021-10-22 | 中国移动通信集团浙江有限公司 | Portrait anti-counterfeiting detection method and device and computing equipment |
| CN114066894A (en) * | 2022-01-17 | 2022-02-18 | 深圳爱莫科技有限公司 | Detection method for display image reproduction, storage medium and processing equipment |
| CN116704209A (en) * | 2023-08-08 | 2023-09-05 | 山东顺发重工有限公司 | Quick flange contour extraction method and system |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112258481A (en) * | 2020-10-23 | 2021-01-22 | 北京云杉世界信息技术有限公司 | Portal photo reproduction detection method |
| CN113450272B (en) * | 2021-06-11 | 2024-04-16 | 广州方图科技有限公司 | Image enhancement method based on sinusoidal variation and application thereof |
| CN113592801A (en) * | 2021-07-23 | 2021-11-02 | 浙江大华技术股份有限公司 | Method and device for detecting stripe interference of video image |
| CN115330826B (en) * | 2022-08-11 | 2024-02-20 | 创新奇智(青岛)科技有限公司 | Linear detection method, linear detection device, electronic equipment and computer readable storage medium |
| CN116342656B (en) * | 2023-03-29 | 2024-04-26 | 华北电力大学 | Space-time image speed measurement method and device based on self-adaptive edge detection |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060067588A1 (en) * | 2004-09-28 | 2006-03-30 | Casio Computer Co., Ltd. | Imaging apparatus, image processing method for imaging apparatus and recording medium |
| CN102521614A (en) * | 2011-12-20 | 2012-06-27 | 中山大学 | Identification method of coping digital image |
| JP2013251005A (en) * | 2013-09-03 | 2013-12-12 | Tateyama System Laboratory Ltd | Image correction method |
| CN104598933A (en) * | 2014-11-13 | 2015-05-06 | 上海交通大学 | Multi-feature fusion based image copying detection method |
| CN104867149A (en) * | 2015-05-16 | 2015-08-26 | 吉林大学 | Duplication image identification algorithm based on local plane linearity point |
| CN105117729A (en) * | 2015-05-11 | 2015-12-02 | 杭州金培科技有限公司 | Method and device for recognizing copied image |
| US20170074808A1 (en) * | 2015-09-11 | 2017-03-16 | Thales | Test pattern and method for calibrating an x-ray imaging device |
| WO2018107939A1 (en) * | 2016-12-14 | 2018-06-21 | 国家海洋局第二海洋研究所 | Edge completeness-based optimal identification method for image segmentation |
| CN108460649A (en) * | 2017-02-22 | 2018-08-28 | 阿里巴巴集团控股有限公司 | A kind of image-recognizing method and device |
| CN108549836A (en) * | 2018-03-09 | 2018-09-18 | 通号通信信息集团有限公司 | Reproduction detection method, device, equipment and the readable storage medium storing program for executing of photo |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016018395A1 (en) * | 2014-07-31 | 2016-02-04 | Hewlett-Packard Development Company, L.P. | Document region detection |
| KR102354458B1 (en) * | 2015-08-19 | 2022-01-21 | 삼성전자주식회사 | Display apparatus and controlling method thereof |
| CN108171689B (en) * | 2017-12-21 | 2021-12-14 | 深圳大学 | Identification method and device for copying display screen image and storage medium |
| CN109859227B (en) * | 2019-01-17 | 2023-07-14 | 平安科技（深圳）有限公司 | Method and device for detecting flip image, computer equipment and storage medium |
-
2019
- 2019-01-17 CN CN201910043484.0A patent/CN109859227B/en active Active
- 2019-11-12 WO PCT/CN2019/117396 patent/WO2020147398A1/en active Application Filing
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060067588A1 (en) * | 2004-09-28 | 2006-03-30 | Casio Computer Co., Ltd. | Imaging apparatus, image processing method for imaging apparatus and recording medium |
| CN102521614A (en) * | 2011-12-20 | 2012-06-27 | 中山大学 | Identification method of coping digital image |
| JP2013251005A (en) * | 2013-09-03 | 2013-12-12 | Tateyama System Laboratory Ltd | Image correction method |
| CN104598933A (en) * | 2014-11-13 | 2015-05-06 | 上海交通大学 | Multi-feature fusion based image copying detection method |
| CN105117729A (en) * | 2015-05-11 | 2015-12-02 | 杭州金培科技有限公司 | Method and device for recognizing copied image |
| CN104867149A (en) * | 2015-05-16 | 2015-08-26 | 吉林大学 | Duplication image identification algorithm based on local plane linearity point |
| US20170074808A1 (en) * | 2015-09-11 | 2017-03-16 | Thales | Test pattern and method for calibrating an x-ray imaging device |
| WO2018107939A1 (en) * | 2016-12-14 | 2018-06-21 | 国家海洋局第二海洋研究所 | Edge completeness-based optimal identification method for image segmentation |
| CN108460649A (en) * | 2017-02-22 | 2018-08-28 | 阿里巴巴集团控股有限公司 | A kind of image-recognizing method and device |
| CN108549836A (en) * | 2018-03-09 | 2018-09-18 | 通号通信信息集团有限公司 | Reproduction detection method, device, equipment and the readable storage medium storing program for executing of photo |
Non-Patent Citations (11)
| Title |
|---|
| LU LIU ET AL.: "Improved SIFT-Based Copy-Move Detection Using BFSN Clustering and CFA Features", 《IEEE XPLORE》 * |
| LU LIU ET AL.: "Improved SIFT-Based Copy-Move Detection Using BFSN Clustering and CFA Features", 《IEEE XPLORE》, 29 December 2014 (2014-12-29) * |
| 奚永忠等: "数字图像的真伪检验", 《产业与科技论坛》 * |
| 奚永忠等: "数字图像的真伪检验", 《产业与科技论坛》, no. 02, 15 January 2017 (2017-01-15) * |
| 尹京等: "数码翻拍图像取证算法", 《中山大学学报(自然科学版)》 * |
| 尹京等: "数码翻拍图像取证算法", 《中山大学学报(自然科学版)》, no. 06, 15 November 2011 (2011-11-15) * |
| 尹京等: "数码翻拍图像取证算法", 《中山大学学报(自然科学版)》, no. 06, 30 November 2011 (2011-11-30) * |
| 谢心谦等: "基于深度学习的图像翻拍检测", 《电脑知识与技术》 * |
| 谢心谦等: "基于深度学习的图像翻拍检测", 《电脑知识与技术》, no. 16, 5 June 2017 (2017-06-05) * |
| 鲁光泉等: "基于链码检测的直线段检测方法", 《计算机工程》 * |
| 鲁光泉等: "基于链码检测的直线段检测方法", 《计算机工程》, no. 14, 20 July 2006 (2006-07-20) * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020147398A1 (en) * | 2019-01-17 | 2020-07-23 | 平安科技（深圳）有限公司 | Reproduced image detection method and device, computer device and storage medium |
| CN110348511A (en) * | 2019-07-08 | 2019-10-18 | 创新奇智(青岛)科技有限公司 | A kind of picture reproduction detection method, system and electronic equipment |
| CN110706277A (en) * | 2019-09-12 | 2020-01-17 | 创新奇智(合肥)科技有限公司 | Method for detecting object rotation angle based on chain code histogram |
| CN111461143A (en) * | 2020-03-31 | 2020-07-28 | 珠海格力电器股份有限公司 | Picture copying identification method and device and electronic equipment |
| CN113542142A (en) * | 2020-04-14 | 2021-10-22 | 中国移动通信集团浙江有限公司 | Portrait anti-counterfeiting detection method and device and computing equipment |
| CN113542142B (en) * | 2020-04-14 | 2024-03-22 | 中国移动通信集团浙江有限公司 | Portrait anti-fake detection method and device and computing equipment |
| CN112115945A (en) * | 2020-09-25 | 2020-12-22 | 有米科技股份有限公司 | Method and device for extracting dynamic and static subtitles |
| CN114066894A (en) * | 2022-01-17 | 2022-02-18 | 深圳爱莫科技有限公司 | Detection method for display image reproduction, storage medium and processing equipment |
| CN116704209A (en) * | 2023-08-08 | 2023-09-05 | 山东顺发重工有限公司 | Quick flange contour extraction method and system |
| CN116704209B (en) * | 2023-08-08 | 2023-10-17 | 山东顺发重工有限公司 | Quick flange contour extraction method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109859227B (en) | 2023-07-14 |
| WO2020147398A1 (en) | 2020-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109859227A (en) | Reproduction image detecting method, device, computer equipment and storage medium | |
| US10885644B2 (en) | Detecting specified image identifiers on objects | |
| US9818023B2 (en) | Enhanced face detection using depth information | |
| CN106778586B (en) | Off-line handwritten signature identification method and system | |
| US10635946B2 (en) | Eyeglass positioning method, apparatus and storage medium | |
| CN102360421B (en) | Face identification method and system based on video streaming | |
| EP2866170B1 (en) | Image processing device and image processing method | |
| US20170357869A1 (en) | Mobile document detection and orientation based on reference object characteristics | |
| CN105956578A (en) | Face verification method based on identity document information | |
| US10586321B2 (en) | Automatic detection, counting, and measurement of lumber boards using a handheld device | |
| CN111144366A (en) | Strange face clustering method based on joint face quality assessment | |
| US20080193020A1 (en) | Method for Facial Features Detection | |
| CN110781885A (en) | Text detection method, device, medium and electronic equipment based on image processing | |
| Guerreiro et al. | Connectivity-enforcing Hough transform for the robust extraction of line segments | |
| CN111626163B (en) | Human face living body detection method and device and computer equipment | |
| JP2002507035A (en) | How to authenticate the validity of an image recorded for personal identification | |
| CN108961262B (en) | Bar code positioning method in complex scene | |
| US11216905B2 (en) | Automatic detection, counting, and measurement of lumber boards using a handheld device | |
| US20220198723A1 (en) | Image enhancement method and image enhancement apparatus | |
| CN109840458A (en) | A kind of fingerprint identification method and fingerprint collecting equipment | |
| CN104966283A (en) | Imaging layered registering method | |
| CN111027637A (en) | Character detection method and computer readable storage medium | |
| US20150112853A1 (en) | Online loan application using image capture at a client device | |
| CN107145820B (en) | Binocular positioning method based on HOG characteristics and FAST algorithm | |
| WO2023028947A1 (en) | Palm vein non-contact three-dimensional modeling method and apparatus, and authentication method |
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 |