CN109859227B - Method and device for detecting flip image, computer equipment and storage medium - Google Patents
Method and device for detecting flip image, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109859227B CN109859227B CN201910043484.0A CN201910043484A CN109859227B CN 109859227 B CN109859227 B CN 109859227B CN 201910043484 A CN201910043484 A CN 201910043484A CN 109859227 B CN109859227 B CN 109859227B
- Authority
- CN
- China
- Prior art keywords
- image
- edge
- straight line
- initial
- straight lines
- 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.)
- Active
Links
Images
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 method, a device, computer equipment and a storage medium for detecting a flip image, wherein the method comprises the following steps: acquiring a test image; performing edge detection on the test image by adopting an edge detection algorithm to obtain an edge image; performing linear extraction on the edge image through a linear detection algorithm to obtain an initial linear; acquiring a flap line from the initial line by adopting a line dense algorithm; and calculating the ratio value of the number of the turnup straight lines to the number of the initial straight lines, and determining that the test image is a suspected turnup image if the ratio value of the straight lines is larger than or equal to a preset ratio value threshold value of the straight lines as the ratio value of the straight lines. By adopting the method, the detection efficiency of the image reproduction is improved, and the image reproduction detection accuracy is improved by adopting the image reproduction detection method to carry out image reproduction detection on the image to be identified.
Description
Technical Field
The present invention relates to the field of image recognition, and in particular, to a method and apparatus for detecting a flipped image, a computer device, and a storage medium.
Background
With the continuous improvement of image display technology, people can acquire high-quality flip images through a plurality of methods, more and more lawbreakers begin to use the images for illegal paths, and the current image evidence obtaining system often has difficulty in detecting whether the images belong to the flip. In view of the potential hazard of the flip image to social order and public safety, research on detection of the flip image becomes an important subject. For example, digital images are now widely used in online business applications, such as online account opening using identification card images, while illicit users can open accounts by flipping evidence information found on the network. Therefore, an accurate method for identifying a flipped image is needed to identify the image uploaded by the user, so as to prevent illegal users from submitting forged image materials by using screen flipping.
Conventionally, the detection and identification of the image are completed by adopting feature extraction and neural network classification calculation, the architecture is overweight, training data is slower, the detection accuracy is lower, the image reproduction detection efficiency is low, and the accuracy of image reproduction identification is affected.
Disclosure of Invention
The embodiment of the invention provides a method, a device, computer equipment and a storage medium for detecting a flipped image, which are used for solving the problem of low image flipped detection efficiency.
In addition, the embodiment of the invention also provides a method, a device, computer equipment and a storage medium for detecting the image in the case of image reproduction, so as to solve the problem of low accuracy of image reproduction detection and identification.
A method of detecting a flip image, comprising:
acquiring a test image;
performing edge detection on the test image by adopting an edge detection algorithm to obtain an edge image;
performing linear extraction on the edge image through a linear detection algorithm to obtain an initial linear;
acquiring a flap line from the initial line by adopting a line dense algorithm;
and calculating the ratio value of the number of the turnup straight lines to the number of the initial straight lines, and determining that the test image is a suspected turnup image if the ratio value of the straight lines is larger than or equal to a preset ratio value threshold value of the straight lines as the ratio value of the straight lines.
A flip image detection apparatus comprising:
the test image acquisition module is used for acquiring a test image;
the edge image acquisition module is used for carrying out edge detection on the test image by adopting an edge detection algorithm to obtain an edge image;
the initial straight line acquisition module is used for carrying out straight line extraction on the edge image through a straight line detection algorithm to obtain an initial straight line;
The overturning straight line acquisition module is used for acquiring overturning straight lines from the initial straight lines by adopting a straight line dense algorithm;
the suspected flap image acquisition module is used for calculating the ratio value of the number of the flap straight lines to the number of the initial straight lines, and determining that the test image is a suspected flap image if the ratio value of the straight lines is larger than or equal to a preset ratio value threshold value of the straight lines.
A method of detecting a flip image, comprising:
acquiring an image to be identified, and dividing the image to be identified into N sub-images according to a preset partitioning mode;
detecting each sub-image by adopting a flip image detection method to obtain the number M of the suspected flip images, wherein M is a positive integer;
and calculating the ratio value of the number M of the suspected flip images to the number N of the sub-images to serve as the sub-image ratio value, and if the sub-image ratio value is larger than or equal to a preset sub-image ratio value threshold, determining that the image to be identified is a flip image.
A flip image detection apparatus comprising:
the sub-image acquisition module is used for acquiring an image to be identified and dividing the image to be identified into N sub-images according to a preset blocking mode;
A suspected flap image quantity obtaining module, configured to detect each sub-image by using the flap image detection method according to any one of claims 1 to 4, to obtain a suspected flap image quantity M, where M is a positive integer;
and the image to be identified is a flap image if the sub-image proportion value is greater than or equal to a preset sub-image proportion value threshold value.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the above-described method of detecting a flipped image when executing the computer program or the processor implementing the above-described method of detecting a flipped image when executing the computer program.
A computer-readable storage medium storing a computer program which, when executed by a processor, implements the above-described method of detecting a tapped image, or which, when executed by the processor, implements the above-described method of detecting a tapped image.
In the method, the device, the computer equipment and the storage medium for detecting the flip image, the test image is firstly obtained, then the edge detection algorithm is adopted to carry out edge detection on the test image, so as to obtain the edge image, and more accurate flip characteristics can be obtained based on the edge image later; then, carrying out linear extraction on the edge image through a linear detection algorithm to obtain an initial linear, so that the pixel characteristics can be more accurately flipped and shot in the initial linear extraction later; then, a straight line dense algorithm is adopted to obtain a reproduction straight line from the initial straight line, so that the accuracy of reproduction image detection is improved; and finally, calculating the ratio value of the number of the turnup straight lines to the number of the initial straight lines, and determining that the test image is a suspected turnup image when the ratio value of the straight lines is larger than or equal to a preset ratio value threshold value as the ratio value of the straight lines, wherein the algorithm is simple, and the efficiency of detecting the turnup image is greatly improved.
In the method, the device, the computer equipment and the storage medium for detecting the flip image, the image to be identified is firstly obtained, the image to be identified is divided into N sub-images according to the preset block mode, and the sub-image blocks contain richer local information, so that the accuracy of detecting the follow-up flip image is improved. And then detecting each sub-image by adopting a reproduction image detection method to obtain the number M of suspected reproduction images. And finally, calculating the ratio value of the number M of suspected flip images to the number N of sub-images, wherein the ratio value is used as the sub-image ratio value, and if the sub-image ratio value is larger than or equal to a preset sub-image ratio value threshold value, the image to be identified is the flip image. The detection method of the image is high in accuracy, so that the suspected image is high in accuracy, the richer and more accurate local information in the sub-images is fully considered, and the accuracy of the image detection of the image is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic view of an application environment of a method for detecting a flip image according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating an exemplary method for detecting a flipped image according to an embodiment of the present invention;
FIG. 3 is another exemplary diagram of a method for detecting a flipped image according to an embodiment of the present invention;
FIG. 4 is another exemplary diagram of a method for detecting a flipped image according to an embodiment of the present invention;
FIG. 5 is another exemplary diagram of a method for detecting a flipped image according to an embodiment of the present invention;
FIG. 6 is a schematic block diagram of a device for detecting a flipped image according to an embodiment of the present invention;
FIG. 7 is another schematic block diagram of a device for detecting a flipped image according to an embodiment of the present invention;
FIG. 8 is a diagram illustrating an exemplary method for detecting a flipped image according to an embodiment of the present invention;
FIG. 9 is another schematic block diagram of a flip image detection apparatus provided by an embodiment of the present invention;
Fig. 10 is a schematic diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The method for detecting the flip image can be applied to an application environment as shown in fig. 1, wherein a client communicates with a server through a network, the server receives a test image sent by the client, and an edge detection algorithm is adopted to detect the edge of the test image to obtain an edge image. And then, extracting the straight line of the edge image by a straight line detection algorithm to obtain an initial straight line. And then acquiring the flap line from the initial line by adopting a line dense algorithm. And further calculating the ratio value of the number of the turnup straight lines to the number of the initial straight lines, and determining that the test image is a suspected turnup image when the ratio value of the straight lines is larger than or equal to a preset ratio value threshold value of the straight lines as the ratio value of the straight lines. The clients may be, but are not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server may be implemented by a stand-alone server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 2, the method is applied to the server in fig. 1, and the method includes the following steps:
s10: and obtaining a test image.
The test image is a two-dimensional image acquired by using an image acquisition tool, such as a two-dimensional image of an outdoor scene, an indoor scene, a person and the like, and is used as an image for performing the detection of the flip image. Alternatively, the test image may be an image acquired in real time by the client, or may be an image stored in advance on the client. Specifically, the client sends the test image to the server, and the server acquires the test image.
S20: and carrying out edge detection on the test image by adopting an edge detection algorithm to obtain an edge image.
The edge of the image refers to the part of the image where the brightness change is significant in the local area, and the gray level profile of the area can be generally regarded as a step, i.e. the gray level value changes sharply from one gray level value to another gray level value with a larger difference in the small buffer area. The edge image refers to an edge contour image of the test image. It will be appreciated that the edge portion of the edge image concentrates most of the profile information of the test image itself, and that the determination and extraction of the edge image is very important for the recognition and understanding of the entire image scene, as well as important features on which the image segmentation depends. Edge detection is the measurement, detection and localization of the grayscale variation of a test image. The edge detection algorithm can be a Sobel edge detection algorithm, a Canny edge detection algorithm, a Roberts edge detection algorithm and the like.
It should be noted that, the edge detection in this step is used to obtain the flipped pixel feature of the image, and the key point of obtaining the flipped pixel feature is to obtain the actual edge image of the test image. It can be understood that the Canny edge detection algorithm uses two different thresholds to detect the strong edge and the weak edge respectively, and only when the weak edge is connected with the strong edge, the weak edge is included in the output image, so that the Canny edge detection algorithm can detect the real weak edge, is an effective edge detection algorithm, can greatly improve the efficiency of the edge detection algorithm, and meanwhile, the phenomenon that the test image becomes blurred due to the flip-flop is shown more obviously at the edge of the image. In this embodiment, an edge detection algorithm is used to perform edge detection on the test image, so as to obtain an edge image, so that more accurate flipped pixel characteristics can be obtained based on the edge image later.
S30: and extracting the straight line of the edge image by a straight line detection algorithm to obtain an initial straight line.
The straight line detection algorithm refers to a detection method for extracting straight lines existing in an image, and the straight line detection algorithm comprises but is not limited to a Hough straight line detection algorithm, a Freeman straight line detection algorithm or an inchworm creeping algorithm. The initial straight line is a straight line obtained by connecting pixel points meeting the straight line condition in the edge image, and a basis is provided for extracting the pixel characteristics of the test image in a flip mode. Specifically, a straight line formed by pixels meeting a straight line condition existing in the connected edge image is extracted as an initial straight line by a straight line detection algorithm. The pixels meeting the linear condition are pixels with equal slopes between every two pixels.
It can be appreciated that, because the image is obtained by performing the image reproduction on the real image through scanning, printing or other devices with shooting functions, the characteristics of the display medium and the scene differences in the image reproduction process are considered in the process of performing the image reproduction on the real image, so that the image reproduction is different from the real image. Such as: and the surface gradient value of the flipped image is abnormal compared with the real image, so that the initial straight line distribution in the flipped image is changed, and therefore, the initial straight line in the edge image is extracted, and the pixel characteristics of the flipped image can be extracted more accurately in the initial straight line.
S40: and acquiring the flap line from the initial line by adopting a line dense algorithm.
The flipped straight line refers to a straight line which is obtained by performing straight line extraction on an initial straight line by adopting a straight line dense algorithm and is provided with flipped pixel characteristics and used for distinguishing flipped pixel characteristics of a flipped image and non-flipped image (real image). Specifically, a line dense algorithm is adopted in the initial line to extract a flipped line with flipped pixel characteristics.
The straight line dense algorithm is an algorithm for extracting straight lines meeting the rule of dense concentration and parallelism of the straight lines, and is used for extracting the characteristic of the flipped pixels in the straight lines. The flipping straight line refers to a straight line meeting the criterion of the straight line dense algorithm, namely the straight line is densely concentrated, the criterion parallel to the straight line comprises that the slope difference value of two straight lines is smaller than 1 degree (degree), and the distance between two adjacent parallel straight lines is smaller than a preset distance threshold. In the line dense algorithm, the initial line needs to satisfy both the conditions of "parallel" and "dense concentration". For the "parallel" condition, i.e., after the slope values of the two initial straight lines are written into the initial straight line pixel points, if the slope difference is smaller than 1 ° (degree), the initial straight lines are parallel, i.e., the "parallel" condition is satisfied. For the condition of "dense concentration", i.e., the distance between two parallel straight lines (initial straight lines) is calculated, two straight lines satisfying the distance less than the preset distance threshold are determined as satisfying the condition of "dense concentration", i.e., the flip straight lines. If the linear equation of the initial line L1 is k 1 x-d 1 =y, the linear equation of the initial line L2 is k 2 x-d 2 =y, and the initial straight line L1 is adjacent to the initial straight line L2, the formula of the criterion is as follows:
wherein k is 1 Is the slope, k of the initial straight line L1 2 Is the slope of the initial straight line L2, d 1 Is the intercept of the initial straight line L1, d 2 For the intercept of the initial straight line L2, α is a preset distance threshold.
Specifically, calculating the slope of the initial straight line, and judging the slope of the initial straight line by a judging criterion of a straight line dense algorithm, namely judging whether any two adjacent initial straight lines meet the judging criterion formula, and if so, determining the initial straight line as a flap straight line. The initial straight line meeting the criterion of the straight line dense algorithm is taken as the reproduction straight line, and as the actual image and the reproduction image are different in pixel block characteristics, the straight line is determined to be the reproduction straight line when the initial straight line meets the dense concentration and parallelism of the straight lines, so that the confirmation of the reproduction pixel characteristics is more accurate. And moreover, a line dense algorithm is adopted to acquire the reproduction line from the initial line, so that the accuracy of reproduction image detection is improved.
S50: and calculating the ratio value of the number of the turnup straight lines to the number of the initial straight lines, and determining that the test image is a suspected turnup image if the ratio value of the straight lines is larger than or equal to a preset ratio value threshold value of the straight lines as the ratio value of the straight lines.
The straight line proportion value refers to a proportion value of the number of the flipping straight lines to the number of the initial straight lines. For example, if the number of the flipping straight lines is a and the number of the initial straight lines is b, the ratio of the straight lines is
Because the flipping straight line is extracted from the initial straight line, a is less than or equal to b, namely the value range of the straight line proportion value is [0,1]. Specifically, the number of the flap lines and the number of the initial lines may be calculated by an accumulator to obtain the number of the flap lines and the number of the initial lines. The preset linear proportion value threshold is a critical value of the linear proportion value of the detection of the flipped image and the real image and is used for determining the limit of the test image as the flipped image or the real image. The suspected flap image is an image with the feature quantity of the flap pixels reaching a preset condition, namely an image with the linear proportion value reaching a threshold value. For example, an initial test imageThe number of the straight lines is 100, the number of the turnup straight lines is 20, the preset straight line proportion value threshold is 10%, the proportion value of the number of the turnup straight lines and the number of the initial straight lines is 20% and is larger than the preset straight line proportion value threshold by 10%, and therefore the test image is a suspected turnup image.
It should be noted that, the preset linear scale threshold values of the images with different resolutions are different and may be set according to practical application requirements, which is not limited herein.
In step S50, a ratio value of the number of the flipped straight lines to the number of the initial straight lines is calculated and used as a straight line ratio value, and the test image with the straight line ratio value greater than or equal to the preset straight line ratio value threshold value is confirmed to be the suspected flipped image. The test image is detected according to the quantity and the proportion relation of the characteristics of the turnup pixels, so that the algorithm is simple, and the efficiency of detecting the turnup image is greatly improved.
In the embodiment, a test image is firstly obtained, then edge detection is carried out on the test image by adopting an edge detection algorithm to obtain an edge image, so that more accurate reproduction characteristics can be obtained based on the edge image later; then, carrying out linear extraction on the edge image through a linear detection algorithm to obtain an initial linear, so that the pixel characteristics can be more accurately flipped and shot in the initial linear extraction later; then, a straight line dense algorithm is adopted to obtain a reproduction straight line from the initial straight line, so that the accuracy of reproduction image detection is improved; and finally, calculating the ratio value of the number of the turnup straight lines to the number of the initial straight lines, and determining that the test image is a suspected turnup image when the ratio value of the straight lines is larger than or equal to a preset ratio value threshold value as the ratio value of the straight lines, wherein the algorithm is simple, and the efficiency of detecting the turnup image is greatly improved.
In one embodiment, as shown in fig. 3, in step S20, edge detection is performed on a test image by using an edge detection algorithm to obtain an edge image, including:
s21: and smoothing the test image by using a Gaussian filter to obtain a smoothed image.
Wherein the gaussian filter is a filter for removing gaussian noise of an image. The smoothing process refers to an operation of removing noise from the test image, and is used for reducing interference of the noise on the test image. Specifically, smoothing processing can be performed by using an image mean smoothing filter gaussian blur () function in Python, so that the information of the smoothed image is more abundant, for example:
blur＝cv2.GaussianBlur(img，(5，5)，0)；
where img is the test image name, blur is the smoothed image name, (5, 5) is the window size of 5×5,0 is the standard deviation of the Gaussian function. It will be appreciated that the above values are merely exemplary and that the values may be specifically set and adjusted according to the actual circumstances.
S22: and calculating the gradient amplitude and direction of the smooth image by adopting the first-order partial derivative finite difference to obtain gradient information.
The gradient information refers to the gradient amplitude and direction of the smoothed image. The gradient amplitude and direction of the smooth image are calculated by adopting the first-order partial derivative finite difference, which comprises the following calculation processes:
First, the gradient of each pixel is obtained, and if the pixel of the smoothed image is (x, y), the gradient calculation formula is as follows:
wherein G is x (x, y) is the horizontal gradient magnitude of pixel (x, y) of the smoothed image, where G y (x, y) is the vertical gradient magnitude of the pixel (x, y) of the smoothed image, and H (x, y) is the gray value of the pixel (x, y).
The pixel gradient size of the smoothed image is then calculated using the following formula:
where G (x, y) is the gradient magnitude of the smoothed image.
Finally, the gradient direction of the smoothed image is calculated using the following formula:
where α (x, y) is the direction angle of the gradient direction of the smoothed image.
It will be appreciated that the gradient direction α (x, y) corresponds to the position in the smoothed image where the gray scale intensity variation is strongest. The strongest change refers to the gradient direction. First derivatives (image gradients) in the horizontal and vertical directions are calculated on the smoothed image using first-order partial derivative finite differences to subsequently determine gradients and directions of the boundary from the gradient information.
S23: and carrying out non-maximum suppression on the gradient information to obtain K edge points, wherein K is a positive integer.
Among them, non-maximum suppression is to sharpen the boundary of smooth image blur (sharp). I.e. the maximum value of the gradient intensity at each pixel point is preserved, while the other values are deleted. Specifically, after gradient information is obtained, non-maximum suppression is performed on the gradient information, so as to obtain K pixel points with the largest pixel value in the same gradient direction, namely K edge points, where K refers to the number of edge points, and the value of K is determined by the gradient information, which is not limited herein.
S24: and connecting strong edge points in the K edge points by using a double-threshold algorithm to obtain an edge image.
The double-threshold algorithm is an algorithm for setting an upper threshold and a lower threshold to select pixel points. Specifically, two thresholds are set: maxVal (upper threshold) and minVal (lower threshold). And comparing each edge point with maxVal (upper threshold) and minVal (lower threshold) respectively, namely when the gray gradient of the edge point is larger than maxVal, confirming that the edge point is a strong edge point, and when the gray gradient of the edge point is smaller than minVal, confirming that the edge point is a weak edge point, and deleting the weak edge point. When the gray gradient of the edge point is smaller than or equal to maxVal and larger than or equal to minVal, judging whether the edge point is connected with a strong edge point, if the edge point is connected with the strong edge point, confirming that the edge point is the strong edge point, otherwise, confirming that the edge point is a weak edge point, and deleting the weak edge point. And further screening out strong edge points, and connecting each strong edge point to form an edge image. Therefore, the strong edge point and the weak edge point can be distinguished through the double-threshold algorithm, so that the detected edge image is more accurate.
In the embodiment, a Gaussian filter is used for smoothing a test image to obtain a smooth image, so that the interference of noise on the test image is reduced, and the information of the smooth image is more abundant; then, calculating the gradient amplitude and direction of the smooth image by adopting a first-order partial derivative finite difference to obtain gradient information so as to determine the gradient and direction of the boundary according to the gradient information; performing non-maximum suppression on the gradient information to obtain K edge points; and finally, connecting strong edge points in the K edge points by using a double-threshold algorithm to obtain an edge image, and distinguishing the strong edge points from the strong edge points by using the double-threshold algorithm so that the detected edge image is more accurate.
In one embodiment, as shown in fig. 4, in step S30, an edge image is extracted by a line detection algorithm to obtain an initial line, which specifically includes the following steps:
s31: and converting the edge image into a binary image to obtain a binary edge image.
The binary image is an image in which the gray value of each pixel is black or white, and the gray value of the binary image has no intermediate transition, and is used for describing the outline of characters or images. The binary edge image refers to an edge image in binary form. Specifically, the contours of the edge images are described by binary form images so that the flipped pixel features of the test image are subsequently extracted by the contours of the test image. It will be appreciated that the outline of the image contains more information about the image, and therefore, by converting the edge image into a binary form of the image, more accurate flipped pixel features are advantageously extracted.
S32: and acquiring pixel data of the binary edge image.
The pixel data refers to data information of pixels of the binary edge image, such as brightness values of the pixels. Specifically, the acquiring of the pixel data of the binary edge image may be acquired by a pixel acquiring tool in the Opencv library, or may be acquired by a pixel acquiring tool in the matlab, which may be selected according to actual requirements, which is not limited herein.
In this embodiment, the pixel data of the binary edge image is acquired, so that the flipped pixel feature is extracted based on the pixel data.
S33: and carrying out Hough transformation on the pixel data according to a Hough transformation formula to obtain an initial straight line.
Among them, hough Transform (Hough Transform) is a feature extraction technique in image processing, which detects an object having a specific shape by a voting algorithm. The hough transform formula of the straight line refers to a discrimination formula for detecting whether the straight line exists in the edge image through the hough transform principle. The Hough transform formula is shown as follows:
xcosθ+ysinθ＝ρρ≥0，0≤θ≤π；
wherein ρ is the distance from the line L to the origin; θ is the angle between the straight line L and the positive direction of the x-axis, and according to the above formula, the different points on the straight line L are transformed into a cluster of sinusoids intersecting at the (x, y) point in the parameter plane ρ - θ. And calculating a sinusoidal track in the Hough parameter space according to the data points in the image space, carrying out two-dimensional statistics on a parameter plane, and selecting a peak value. The peak value is a parameter of a straight line in the image space, specifically, pixel data is used as input of a Hough transformation formula, and initial straight lines are obtained through Hough transformation, so that straight line detection in the image space is realized, the number of the detected straight lines is accumulated, and the number of the initial straight lines of the edge image is further determined.
Specifically, the HoughLines () function and the HoughLinesP () function are respectively arranged in opencv, so that Hough straight line detection can be realized. Wherein, the HoughLines () function implementation is a standard Hough transform. The standard hough transform is to map an edge image onto its parameter space and determine an initial straight line by computing all edge points. The initial straight line can be determined quickly and accurately.
In this embodiment, the edge image is first converted into a binary image to obtain a binary edge image, which is favorable for extracting more accurate flipped pixel features. Then, pixel data of the binary edge image is acquired to extract the flipped pixel feature. And finally, carrying out Hough transformation on the pixel data according to a Hough transformation formula to obtain an initial straight line, and rapidly and accurately acquiring the initial straight line.
In one embodiment, as shown in fig. 5, in step S30, an edge image is extracted by a line detection algorithm to obtain an initial line, which specifically includes the following steps:
s31': and detecting the chain code of the edge image to obtain the chain code of the edge image, and calculating a histogram of the chain code.
The chain code is also called a freeman code, and is a method for describing a curve or a boundary by using coordinates of a starting point of the curve and direction codes of boundary points, and is used for representing a coding representation of the curve and the boundary of a region. The chain code detection refers to the processing of detecting the chain code of the obtained image after the chain code tracking and the chain code connection are carried out on the image. Wherein the calculation of the histogram of the chain code may be performed according to the definition of the histogram, for example: chain code { a } for edge image i } n (a i E {0,1,2,3,4,5,6,7 }), taking K points before the point and K points after the point to calculate a forward and a backward chain code histogram at the point, and obtaining a chain code histogram.
Specifically, by detecting the boundary line of the edge image, the chain code of the edge image is obtained, and the histogram of the chain code is calculated, so that the straight line parameter is obtained based on the histogram of the chain code.
S32': based on the histogram of the chain code, the linear parameters of the linear chain code are obtained.
The straight line parameter refers to a straight line parameter on a two-dimensional plane, and the straight line parameter comprises 4 parameters of a starting point coordinate and an end point coordinate. Specifically, the maxima of the forward chain code histogram and the backward chain code histogram of the chain code in step S31' are calculated, and when the maxima of the forward chain code histogram and the backward chain code histogram are equal, the chain code is illustrated as a straight line, and then the straight line parameters of the straight line segment where the straight line is located, namely, the start point coordinate and the end point coordinate are extracted.
S33': and taking the straight line determined by the straight line parameter as an initial straight line.
Specifically, a least square method is adopted to estimate the straight line parameters to obtain a set of straight line equations, and the straight line determined by the straight line equations is the initial straight line. By estimating the straight line parameters, the deviation caused by determining the straight line is effectively overcome, the straight line parameter information is effectively utilized, and the fitting precision of the initial straight line is improved.
In this embodiment, first, chain code detection is performed on an edge image to obtain a chain code of the edge image, and a histogram of the chain code is calculated. And then acquiring the linear parameters of the linear chain codes based on the histogram of the chain codes. And finally, taking the straight line determined by the straight line parameters as an initial straight line, effectively overcoming the deviation caused by the determined straight line, effectively utilizing the straight line parameter information and improving the fitting precision of the initial straight line.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.
In one embodiment, a device for detecting a flipped image is provided, where the flipped image detection device corresponds to the flipped image detection method in the above embodiment one by one. As shown in fig. 6, the apparatus for detecting a reproduction image includes a test image acquisition module 10, an edge image acquisition module 20, an initial straight line acquisition module 30, a reproduction straight line acquisition module 40, and a suspected reproduction image acquisition module 50. The functional modules are described in detail as follows:
a test image acquisition module 10 for acquiring a test image;
The edge image obtaining module 20 is configured to perform edge detection on the test image by using an edge detection algorithm to obtain an edge image;
an initial straight line acquisition module 30, configured to perform straight line extraction on the edge image through a straight line detection algorithm, so as to obtain an initial straight line;
a reproduction straight line acquisition module 40, configured to acquire a reproduction straight line from the initial straight line by using a straight line dense algorithm;
the suspected flap image obtaining module 50 is configured to calculate a ratio value of the number of flap lines to the number of initial lines, and determine that the test image is a suspected flap image if the ratio value of the lines is greater than or equal to a preset ratio value threshold.
Preferably, the edge image acquisition module includes a smooth image acquisition unit, a gradient information acquisition unit, an edge point acquisition unit, and an edge image acquisition unit.
A smooth image acquisition unit for performing smoothing processing on the test image by using a Gaussian filter to obtain a smooth image;
the gradient information acquisition unit is used for calculating the gradient amplitude and direction of the smooth image by adopting the first-order partial derivative finite difference to obtain gradient information;
the edge point acquisition unit is used for carrying out non-maximum suppression on the gradient information to obtain K edge points, wherein K is a positive integer;
And the edge image acquisition unit is used for connecting strong edge points in the K edge points by using a double-threshold algorithm to obtain an edge image.
Preferably, the initial straight line acquisition module includes a binary edge image acquisition unit, a pixel data acquisition unit, and an initial straight line acquisition unit.
The binary edge image acquisition unit is used for converting the edge image into a binary image to obtain a binary edge image;
a pixel data acquisition unit for acquiring pixel data of the binary edge image;
and the initial straight line acquisition unit is used for carrying out Hough transformation on the pixel data according to the Hough transformation formula to obtain an initial straight line.
Preferably, as shown in fig. 7, the initial straight line acquisition module 30 further includes a histogram calculation unit 31', a straight line parameter acquisition unit 32', and an initial straight line acquisition unit 33' of the chain code.
A chain code histogram calculation unit 31' for performing chain code detection on the edge image to obtain a chain code of the edge image, and calculating a histogram of the chain code;
a straight line parameter obtaining unit 32' for obtaining straight line parameters of the straight line chain code based on the histogram of the chain code;
an initial straight line acquisition unit 33' for taking the straight line determined by the straight line parameter as an initial straight line.
In an embodiment, a method for detecting a flipped image is provided, and the method for detecting a flipped image may also be applied in an application environment as shown in fig. 1, where a client communicates with a server through a network. The server receives an image to be identified sent by the client and divides the image to be identified into N sub-images according to a preset blocking mode. And detecting each sub-image by adopting a flip image detection method to obtain the number M of suspected flip images, and finally calculating the ratio value of the number M of suspected flip images to the number N of the sub-images to serve as the sub-image ratio value, wherein if the sub-image ratio value is larger than or equal to a preset sub-image ratio value threshold, the image to be identified is the flip image. The clients may be, but are not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server may be implemented by a stand-alone server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 8, the method is applied to the server in fig. 1, and includes the following steps:
s60: and acquiring an image to be identified, and dividing the image to be identified into N sub-images according to a preset blocking mode.
The image to be identified refers to an image which needs to be identified whether the image is a flip image or not. The sub-image is each block of image after dividing the image to be recognized. The preset partitioning method refers to a preset partitioning method, for example, the image to be identified is partitioned according to a partitioning method of 9×9 equal partition, and the number of sub-images is 81, that is, n=81. For another example, the image to be identified is segmented according to a block dividing mode of dividing the image into horizontal X equally and vertical Y equally, wherein X is a positive integer, Y is a positive integer, and at this time, the number of sub-images is x×y, that is, n=x×y. In this embodiment, the image to be identified is divided into N sub-images according to a preset partitioning manner, so that the subsequent detection of the flip image is performed based on each sub-image block, and it can be understood that the sub-image block contains richer local information, so that the accuracy of the subsequent detection of the flip image is improved.
The preset blocking mode can be set according to the resolution of the image to be detected, and is not limited herein.
S70: and detecting each sub-image by adopting a reproduction image detection method to obtain the number M of suspected reproduction images, wherein M is a positive integer.
Specifically, a method for detecting the flipped image is adopted for each sub-image block, if the sub-image is a suspected flipped image, the number of the suspected flipped images is increased by 1, and if the sub-image is not the suspected flipped image, the number of the suspected flipped images is unchanged. For example, if there are 27 suspected flip images in the 81 sub-image blocks, then m=27.
S80: and calculating the ratio value of the number M of suspected turnup images and the number N of sub-images to serve as the sub-image ratio value, and if the sub-image ratio value is larger than or equal to a preset sub-image ratio value threshold value, determining that the image to be identified is the turnup image.
Specifically, the preset sub-image scale value threshold is a preset critical value for identifying whether the image to be identified is a sub-image scale value of a flip image. In a specific embodiment, the preset sub-image scale value threshold is 20%, and the sub-image scale value is 27/81=30%, so the image to be identified is a flip image. In the embodiment, the number of the sub-images and the number of the suspected flip images are counted, and whether the identification image is the flip image or not is judged according to the ratio value of the sub-images and the suspected flip image, and the accuracy of detecting the suspected flip image is high due to the fact that the accuracy of detecting the flip image is high, the richer and more accurate local information in the sub-images is fully considered, and the accuracy of detecting the image flip image is improved.
In this embodiment, the image to be identified is first obtained, and the image to be identified is divided into N sub-images according to a preset partitioning manner, where the sub-image blocks include richer local information, so that the accuracy of detecting the subsequent flip image is improved. And then detecting each sub-image by adopting a reproduction image detection method to obtain the number M of suspected reproduction images. And finally, calculating the ratio value of the number M of suspected flip images to the number N of sub-images, wherein the ratio value is used as the sub-image ratio value, and if the sub-image ratio value is larger than or equal to a preset sub-image ratio value threshold value, the image to be identified is the flip image. The detection method of the image is high in accuracy, so that the suspected image is high in accuracy, the richer and more accurate local information in the sub-images is fully considered, and the accuracy of the image detection of the image is improved.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.
In one embodiment, a device for detecting a flipped image is provided, where the flipped image detection device corresponds to the flipped image detection method in the above embodiment one by one. As shown in fig. 9, the apparatus for detecting a reproduction image includes a sub-image acquisition module 60, a number of suspected reproduction images acquisition module 70, and a reproduction image identification module 80. The functional modules are described in detail as follows:
the sub-image obtaining module 60 is configured to obtain an image to be identified, and divide the image to be identified into N sub-images according to a preset partitioning manner;
the number of suspected flip images obtaining module 70 is configured to detect each sub-image by using a flip image detection method, and obtain the number M of suspected flip images, where M is a positive integer;
the flipped image recognition module 80 is configured to calculate a ratio value of the number M of suspected flipped images to the number N of sub-images, and if the sub-image ratio value is greater than or equal to a preset sub-image ratio value threshold, the image to be recognized is a flipped image.
For specific limitation of the apparatus for detecting a roll-over, reference is made to the above limitation of the method for detecting a roll-over image, and no further description is given here. The above-mentioned each module in the apparatus for detecting a roll-over can be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 10. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing the test image and the image to be identified. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of detecting a flipped image.
In one embodiment, a computer device is provided that includes a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the method for detecting a flip image in the above embodiments when executing the computer program.
In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, which when executed by a processor, implements the method for detecting a flipped image in the above embodiment.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.
Claims (10)
1. A method for detecting a flip image, the method comprising:
acquiring a test image;
performing edge detection on the test image by adopting an edge detection algorithm to obtain an edge image;
Performing linear extraction on the edge image through a linear detection algorithm to obtain an initial linear;
acquiring a flap line from the initial line by adopting a line dense algorithm; the straight line dense algorithm comprises a discriminant criterion of the flap straight line; the discriminant criteria include: the difference of the slopes of the two straight lines is smaller than 1 DEG, and the distance between the two adjacent parallel straight lines is smaller than a preset distance threshold;
and calculating the ratio value of the number of the turnup straight lines to the number of the initial straight lines, and determining that the test image is a suspected turnup image if the ratio value of the straight lines is larger than or equal to a preset ratio value threshold value of the straight lines as the ratio value of the straight lines.
2. A method for detecting a flip image as defined in claim 1, wherein said edge detection of said test image using an edge detection algorithm to obtain an edge image comprises:
smoothing the test image by using a Gaussian filter to obtain a smoothed image;
calculating the gradient amplitude and direction of the smooth image by adopting a first-order partial derivative finite difference to obtain gradient information;
performing non-maximum suppression on the gradient information to obtain K edge points, wherein K is a positive integer;
And connecting strong edge points in the K edge points by using a double-threshold algorithm to obtain the edge image.
3. The method for detecting a flip image as claimed in claim 2, wherein the step of extracting the edge image by a line detection algorithm to obtain an initial line comprises:
converting the edge image into a binary image to obtain a binary edge image;
acquiring pixel data of the binary edge image;
and carrying out Hough transformation on the pixel data according to a Hough transformation formula to obtain the initial straight line.
4. The method for detecting a flip image as claimed in claim 1, wherein the step of extracting the edge image by a line detection algorithm to obtain an initial line comprises:
performing chain code detection on the edge image to obtain a chain code of the edge image, and calculating a histogram of the chain code;
based on the histogram of the chain code, acquiring a linear parameter of a linear chain code;
and taking the straight line determined by the straight line parameter as the initial straight line.
5. A method for detecting a flip image, the method comprising:
acquiring an image to be identified, and dividing the image to be identified into N sub-images according to a preset partitioning mode;
Detecting each sub-image by adopting the method for detecting the flip-flop images according to any one of claims 1 to 4, and obtaining the number M of the suspected flip-flop images, wherein M is a positive integer;
and calculating the ratio value of the number M of the suspected flip images to the number N of the sub-images to serve as the sub-image ratio value, and if the sub-image ratio value is larger than or equal to a preset sub-image ratio value threshold, determining that the image to be identified is a flip image.
6. A reproduction image detection apparatus, characterized in that the reproduction image detection apparatus includes:
the test image acquisition module is used for acquiring a test image;
the edge image acquisition module is used for carrying out edge detection on the test image by adopting an edge detection algorithm to obtain an edge image;
the initial straight line acquisition module is used for carrying out straight line extraction on the edge image through a straight line detection algorithm to obtain an initial straight line;
the overturning straight line acquisition module is used for acquiring overturning straight lines from the initial straight lines by adopting a straight line dense algorithm; the straight line dense algorithm comprises a discriminant criterion of the flap straight line; the discriminant criteria include: the difference of the slopes of the two straight lines is smaller than 1 DEG, and the distance between the two adjacent parallel straight lines is smaller than a preset distance threshold;
The suspected flap image acquisition module is used for calculating the ratio value of the number of the flap straight lines to the number of the initial straight lines, and determining that the test image is a suspected flap image if the ratio value of the straight lines is larger than or equal to a preset ratio value threshold value of the straight lines.
7. A flip-flop image detection device of claim 6 and wherein said initial straight line acquisition module comprises:
the histogram calculation unit of the chain code is used for carrying out chain code detection on the edge image to obtain the chain code of the edge image, and calculating the histogram of the chain code;
the linear parameter acquisition unit is used for acquiring linear parameters of the linear chain code based on the histogram of the chain code;
and the initial straight line acquisition unit is used for taking the straight line determined by the straight line parameter as the initial straight line.
8. A reproduction image detection apparatus, characterized in that the reproduction image detection apparatus includes:
the sub-image acquisition module is used for acquiring an image to be identified and dividing the image to be identified into N sub-images according to a preset blocking mode;
a suspected-to-flap-image number obtaining module, configured to detect each sub-image by using the flap-image detection method according to any one of claims 1 to 4, to obtain a suspected-to-flap-image number M, where M is a positive integer;
And the image to be identified is a flap image if the sub-image proportion value is greater than or equal to a preset sub-image proportion value threshold value.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the method of detecting a tapped image according to any one of claims 1 to 4 when executing the computer program or the processor implements the method of detecting a tapped image according to claim 5 when executing the computer program.
10. A computer-readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the method of detecting a flip image as claimed in any one of claims 1 to 4, or wherein the computer program when executed by a processor implements the method of detecting a flip image as claimed in claim 5.
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 CN109859227A (en) | 2019-06-07 |
| CN109859227B true 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) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| 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 |
| CN112115945B (en) * | 2020-09-25 | 2024-06-14 | 有米科技股份有限公司 | Dynamic and static caption extraction method and device |
| 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 |
| 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 |
| 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 |
| 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 |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4182937B2 (en) * | 2004-09-28 | 2008-11-19 | カシオ計算機株式会社 | Image capturing apparatus, image processing method for image capturing apparatus, and program |
| 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 |
| US10002434B2 (en) * | 2014-07-31 | 2018-06-19 | Hewlett-Packard Development Company, L.P. | Document region detection |
| CN104598933B (en) * | 2014-11-13 | 2017-12-15 | 上海交通大学 | A kind of image reproduction detection method based on multi-feature fusion |
| CN105117729B (en) * | 2015-05-11 | 2018-11-06 | 杭州集专知识产权代理有限公司 | A kind of method and apparatus of identification reproduction image |
| CN104867149B (en) * | 2015-05-16 | 2017-07-21 | 吉林大学 | Reproduction image authentication method based on part plan Linear Points |
| KR102354458B1 (en) * | 2015-08-19 | 2022-01-21 | 삼성전자주식회사 | Display apparatus and controlling method thereof |
| FR3040867A1 (en) * | 2015-09-11 | 2017-03-17 | Thales Sa | MIRE AND METHOD FOR CALIBRATING AN X-RAY IMAGING SYSTEM |
| 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 |
| CN108171689B (en) * | 2017-12-21 | 2021-12-14 | 深圳大学 | Identification method and device for copying display screen image and storage medium |
| CN108549836B (en) * | 2018-03-09 | 2021-04-06 | 通号通信信息集团有限公司 | Photo copying detection method, device, equipment and readable 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 (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109859227A (en) | 2019-06-07 |
| WO2020147398A1 (en) | 2020-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109859227B (en) | Method and device for detecting flip image, computer equipment and storage medium | |
| CN110414507B (en) | License plate recognition method and device, computer equipment and storage medium | |
| CN110119741B (en) | Card image information identification method with background | |
| CN109492642B (en) | License plate recognition method, license plate recognition device, computer equipment and storage medium | |
| CN110532851B (en) | Finger vein identification method and device, computer equipment and storage medium | |
| CN109447117B (en) | Double-layer license plate recognition method and device, computer equipment and storage medium | |
| CN109903272B (en) | Target detection method, device, equipment, computer equipment and storage medium | |
| CN109255802B (en) | Pedestrian tracking method, device, computer equipment and storage medium | |
| CN113781406B (en) | Scratch detection method and device for electronic component and computer equipment | |
| CN110502977B (en) | Building change classification detection method, system, device and storage medium | |
| CN112464829B (en) | Pupil positioning method, pupil positioning equipment, storage medium and sight tracking system | |
| CN111985427A (en) | Living body detection method, living body detection apparatus, and readable storage medium | |
| CN108665495B (en) | Image processing method and device and mobile terminal | |
| CN111507165A (en) | Face recognition method and device, electronic equipment and computer readable storage medium | |
| US20220245803A1 (en) | Image enhancement processing method, device, equipment, and medium based on artificial intelligence | |
| CN111209908B (en) | Method, device, storage medium and computer equipment for updating annotation frame | |
| CN112465807A (en) | License plate image authenticity identification method, device, equipment and medium | |
| CN112085721A (en) | Damage assessment method, device and equipment for flooded vehicle based on artificial intelligence and storage medium | |
| CN113313092B (en) | Handwritten signature recognition method, and claims settlement automation processing method, device and equipment | |
| CN109447942B (en) | Image ambiguity determining method, apparatus, computer device and storage medium | |
| US10210414B2 (en) | Object detection system and computer program product | |
| Boiangiu et al. | Handwritten documents text line segmentation based on information energy | |
| JP5285401B2 (en) | Face recognition system | |
| CN110751623A (en) | Joint feature-based defect detection method, device, equipment and storage medium | |
| CN116188826A (en) | Template matching method and device under complex illumination condition |
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 |