KR101725041B1 - Method and apparatus for detecting forged image - Google Patents
Method and apparatus for detecting forged image Download PDFInfo
- Publication number
- KR101725041B1 KR101725041B1 KR1020150104105A KR20150104105A KR101725041B1 KR 101725041 B1 KR101725041 B1 KR 101725041B1 KR 1020150104105 A KR1020150104105 A KR 1020150104105A KR 20150104105 A KR20150104105 A KR 20150104105A KR 101725041 B1 KR101725041 B1 KR 101725041B1
- Authority
- KR
- South Korea
- Prior art keywords
- edge
- image
- blur
- target image
- blur index
- Prior art date
Links
Images
Classifications
-
- G06K9/4604—
-
- G06K9/4609—
-
- G06K9/6204—
-
- G06K9/6218—
Abstract
A forged image detection method is provided. A forged image detecting method is characterized in that an edge detector detects an edge point of a check target image and a blur estimator detects a blur of an edge point of the check target image detected by the edge detecting unit The statistical parameter detector models the blur index function using the blur index measured by the blur index measuring section, estimates a statistical parameter based on the modeled blur index function, And the determination unit clusters the statistical parameters estimated by the statistical parameter detection unit and determines a fake area of the verification target image using the group.
Description
The present invention relates to a method and apparatus for detecting a counterfeit image.
Today, image forgery has become easier due to the development of digital cameras and various image processing programs.
Falsified images can be used for crime, etc., and it is important to judge whether the images are forged or not.
At the time of image synthesis, the edge portion of the synthesized region can be blurred to hide the heterogeneous portion of the synthesized region.
The natural blur may be uniformly distributed throughout the image. However, artificial blur can be concentrated in one area.
Judging the falsification area of the image using the blurred part may be an effective method for judging whether the image is falsified or not.
The present invention provides a method for determining whether an image is falsified by using an edge blur index.
According to an aspect of the present invention, there is provided a computer-readable recording medium storing a program for determining whether or not an image is falsified by using a blur index of an edge of an image to be identified.
The technical objects of the present invention are not limited to the technical matters mentioned above, and other technical subjects not mentioned can be clearly understood by those skilled in the art from the following description.
According to another aspect of the present invention, there is provided a method for detecting a fake image, the method comprising: detecting an edge point of an image to be checked; generating a blur estimator; The blur index of the edge of the image to be checked detected by the edge detecting unit is measured and the statistical parameter detector models the blur exponent function using the blur index measured by the blur index measuring unit, The method includes clustering statistical parameters estimated by the statistical parameter detector and determining a forgery area of the image to be checked using the forged region determinator based on the statistical parameters.
In some embodiments, the falsification area determination unit clusters the statistical parameters estimated by the statistical parameter detection unit and uses the falsification area determination unit to determine the falsification area of the verification target image, the determination of the locally dense part of the statistical parameter And determining the area as a forgery area of the target image.
In some embodiments, the edge point of the image to be identified includes a plurality of edge points, and wherein the blur index measuring section measures the blur index includes measuring the blur index for the plurality of edge points can do.
In some embodiments, the edge detector detecting an edge point of the image to be identified includes measuring an edge sample at an edge point of the image to be identified, wherein the blur index is a cost function using the edge sample To a minimum.
According to an aspect of the present invention, there is provided a computer-readable recording medium for detecting an edge point of an image to be checked, measuring a blur index of an edge point of the detected image to be checked, Estimating a statistical parameter based on the modeled blur exponential function, clustering the estimated statistical parameter, and using the calculated blur exponent function to calculate a forgery area of the check target image And the like.
In some embodiments, determining the forgery region of the image to be identified may include determining a locally dense portion of the statistical parameter as a forgery region of the image to be identified.
In some embodiments, the edge point of the image to be identified includes a plurality of edge points, and measuring the blur index of the edge point of the detected image to be checked may include calculating the blur index for the plurality of edge points ≪ / RTI >
In some embodiments, detecting an edge point of the image to be identified includes measuring an edge sample at an edge point of the image to be identified, wherein the blur index is determined by minimizing a cost function using the edge sample Lt; / RTI >
The details of other embodiments are included in the detailed description and drawings.
1 is a block diagram of a fake video detection system according to embodiments of the present invention.
2 is a flowchart of a method for detecting a false image according to embodiments of the present invention.
3 is a flowchart of a blur index measuring method according to embodiments of the present invention.
4 is a diagram for explaining an edge sample detection method according to embodiments of the present invention.
5 is a graph showing the blur exponent function.
6 is a view showing an edge model according to the blur index.
7 is a graph showing a probability density function of the blur index.
8 is a graph showing distribution of statistical parameters before and after blurring.
FIGS. 9A and 9B are diagrams for explaining a result of detection of a false image according to the embodiments of the present invention.
10 is a diagram illustrating a computing system including a program for detecting false images in accordance with embodiments of the present invention.
BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention, and the manner of achieving them, will be apparent from and elucidated with reference to the embodiments described hereinafter in conjunction with the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims. The dimensions and relative sizes of the components shown in the figures may be exaggerated for clarity of description. Like reference numerals refer to like elements throughout the specification and "and / or" include each and every combination of one or more of the mentioned items.
The terminology used herein is for the purpose of illustrating embodiments and is not intended to be limiting of the present invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification. The terms " comprises "and / or" comprising "used in the specification do not exclude the presence or addition of one or more other elements in addition to the stated element.
Unless defined otherwise, all terms (including technical and scientific terms) used herein may be used in a sense commonly understood by one of ordinary skill in the art to which this invention belongs. Also, commonly used predefined terms are not ideally or excessively interpreted unless explicitly defined otherwise.
Referring to FIG. 1, a description will be given of a fake video detection system according to the present invention.
1 is a block diagram of a fake video detection system according to embodiments of the present invention.
The falsification
The falsification
The term " part " used in this embodiment means a hardware component such as software or an FPGA or an ASIC, and 'part' performs certain roles. However, 'minus' is not limited to software or hardware. The " part " may be configured to be in an addressable storage medium and configured to play back one or more processors. Thus, by way of example, and by no means, the terms " component " or " component " means any combination of components, such as software components, object- oriented software components, class components and task components, Subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables. The functions provided in the components and parts may be combined into a smaller number of components and parts or further separated into additional components and parts.
The
Specifically, the
At this time, the edge of the check target image may be composed of an edge point. Further, the number of edge points of the verification target image may be plural. That is, the edge point of the check target image may include a plurality of edge points.
The detection of the edge point of the image to be checked by the
When the
Specifically, the blur
In addition, the blur
Here, the blur index measured by the blur
For example, the blur
At this time, when there are a plurality of edge points of the verification target image, the blur
The
The blur exponential function may, for example, consist of an impulse function and an arbitrary function. In addition, the
The forged
Specifically, the falsification
Hereinafter, a method of detecting a false image of a false image detecting system according to embodiments of the present invention will be described with reference to FIGS. 1 to 8. FIG.
2 is a flowchart of a method of detecting a false image according to embodiments of the present invention.
Referring to FIG. 2, an edge of a check target image is detected (S100). For example, the
Here, the confirmation target image may be a video image for checking whether there is a falsified portion.
The edge of the confirmation target image may be, for example, a part where the brightness rapidly changes in the confirmation target image. In addition, the edge of the confirmation target image may include, for example, a high frequency component.
Therefore, the
In some embodiments, the edge of the image to be identified may be detected using the Canny edge detection method.
When the
The edge point of the check target image may be an object for measuring the blur index to be described below.
Hereinafter, a method of obtaining the edge specimen (Sr) will be described with reference to Figs. 3 and 4. Fig.
FIG. 3 is a flowchart of a blur index measuring method according to an embodiment of the present invention, and FIG. 4 is a view for explaining an edge sample detecting method according to embodiments of the present invention.
Referring to FIG. 3, after detecting an edge of a check target image (S100), a window centered on one edge point is set (S101). For example, the
The window may be, for example, a square shape.
A window may contain one edge point and may be set around the included edge point. However, the embodiments of the present invention are not limited thereto.
For example, one or more edge points may be included in a window. In this case, however, only the edge point in the center of the window can be subjected to blur index measurement, which will be described later.
Next, the edge specimen (Sr) can be measured by measuring the advancing direction of the edge of the confirmation target image at one edge point in the center of the window (S102). For example, the
Referring to FIG. 4, (a) shows a window set around the edge point of the image to be checked, (b) shows the edge direction at the edge point and the edge direction of the image to be checked, and c) is an edge specimen (Sr) graph measured at the edge point of the image to be confirmed.
In step S102, in order to measure the edge specimen (Sr) of the image to be confirmed, it may be necessary to set the direction to display the edge in one-dimensional cross-section.
Referring to FIG. 4B, the edge specimen Sr may be a specimen obtained by extracting values of a normal perpendicular to an inclination direction of edges of an image to be identified.
Specifically, in order to obtain the edge specimen (Sr), the inclination degree ratio, that is, the normal direction of the edge can be measured by obtaining the inclination ratio in the vertical direction and the horizontal direction of the edge of the confirmation object image.
In this case, the direction may be divided into vertical, horizontal, diagonal, or diagonal directions.
The edge specimen (Sr) may have an extended width, for example, one space on the basis of the edge point of the image to be confirmed at the center of the window.
For example, the maximum width (W) may be a total of 13 points, 6 points to the left and right.
In Fig. 4, W may be the maximum width, l the position of the edge point, h the luminosity of the background, and k the variable representing the contrast of the edge.
Referring to (c) of FIG. 4, the edge sample (Sr) is measured under the above-described conditions. Where the x-axis is the edge sample (Sr) and the y-axis is the maximum width (W).
Referring again to FIG. 3, an optimization method is performed on a cost function using the edge specimen Sr (S103). For example, the blur
The cost function may be the difference between the edge specimen Sr of the confirmation target image and the edge model Sa blurred.
For example, the cost function (E) can be defined as: " (1) "
Here, the blurred edge model Sa may be a value that is predicted when the edge of the sample image is blurred.
The edge of the sample image may have the form of a step function.
The blurred edge model Sa can be obtained by gaussian blurring, for example, a step function representing an edge.
For example, the blurred edge model Sa can be defined as Equation (2).
Here, erf is an error function and can be defined as Equation (3).
In Equations (1) and (2),? Can be defined as a blur index. The blur index can be used as a basis for judging a fake area of a check target image.
Here, the blur index can be obtained by the least squares method of the edge sample (Sr) of the image to be confirmed and the edge model (Sa) blurred.
That is, the blur index may be a value that minimizes the cost function using the edge sample Sr.
In order to obtain a blur index that minimizes the cost function, for example, an optimization method may be used.
As an optimization method, for example, a 'Nedler-Mead optimization method' may be used.
The initial parameters to be input in the optimization method may be h 0 , k 0 , l 0 , σ 0 . Where h 0 is the minimum value of the sample, k 0 is the difference between the maximum and minimum values of the sample, l 0 is the center position of the edge, and σ 0 can be any value.
Using the initial parameters, an optimization method can be performed on the cost function to obtain the appropriate parameters (h, k, l, σ).
Referring to FIG. 2 again, the blur index is measured at one edge point of the image to be checked (S110). For example, the blur
At this time, when there are a plurality of edge points of the check target image, the blur index is measured at all edge points (S120). For example, the blur
If the blur index is not measured at all edge points of the check target image, steps S101, S102, S103, and S110 may be repeated for other edge points (S130).
If the blur index is measured at all edge points of the target image, the blur index function is modeled using the blur index (S140). For example, the
In order to model the blur exponential function, the
Given a random variable that is a blur index, the blur exponent function can be estimated from the measured histogram.
For example, the estimated blur exponential function p () can be modeled as an impulse function and an arbitrary function, as in equation (4).
In Equation (4), the impulse function? (?) Having the coefficient k may be composed of values extracted at the edge point of the non-blurred check target image.
Also, in equation (4), any function p b (?) May be a probability density function of the blur index.
Referring to FIG. 5, a graph of the modeled Blur exponent function can be seen.
5 is a graph showing the blur exponent function.
In the graph, the x-axis () is the blur index and the y-axis (p ()) represents the modeled blur exponential function.
At this time, the probability density function p b () of the blur index is
Lt; / RTI >here
Is a value required for statistical parameter estimation to be described later, and can be determined through experiments.6,
Will be described.6 is a view showing an edge model according to blur indexes.
Referring to the graphs of FIG. 6, it can be seen that the edge models change according to the blur index. FIG. 6 exemplarily shows a graph implemented using equation (2) when the parameters h = 0, k = 1, and l = 6.
For example, if σ is a blur exponent, it can diverge when σ = 0, and keep the shape of the step function when 0 <σ <0.22. When σ has a value of 0.22 or more, the graph shape may have a gentle shape as the value increases. If σ exceeds 10, it may not have a step function form.
therefore,
0.22, Can be set to 10.In this case, in the modeled blur exponential function, the probability density function p b (σ) of the blur index can have a range of 0.22 <σ ≤ 10.
The blur index has a non-negative value and can be represented by a continuous probability distribution.
Referring again to FIG. 2, a statistical parameter is estimated based on the modeled Blur exponent function (S150). For example, the
In some embodiments, statistical parameter estimates for the blur index may utilize a " maximum likelihood " method. At this time, the statistical parameter estimation can use the probability density function p b (?) Of the blur index.
In the example described above, the probability density function p b (σ) of the blur index can have a range of 0.22 <σ ≤ 10.
7 is a graph showing a probability density function of the blur index within a range of 0.22 <
Wherein the x-axis blur factor (σ) is, y axis represents the probability density function of the blur index p b (σ).
First, the
For example, when the maximum likelihood method is used, the probability density function p b () of the blur index modeled in step S 140 can be defined as Equation (5).
Here, c, k, and alpha may be statistical parameters.
Specifically, c and k may be shape parameters and may have values greater than zero. Also, alpha may be a scale parameter and may have a value greater than zero.
The likelihood function having the random variable x = (x1, x2, ..., xN ) for obtaining the maximum likelihood may be as shown in Equation (6).
Where θ can be a vector of statistical parameters,
Can be expressed by Equation (7).
here
(C, k,?) Of the probability density function p b (?) Of the blur index, that is, the statistical parameter (c, k,?).A description of the statistical parameters (c, k, alpha) obtained in the above-described manner will be described with reference to Fig.
Fig. 8 is a graph showing distribution of statistical parameters (c, k, alpha) before and after blurring.
Referring to FIG. 8, the x-axis of the graph is the blur index obtained in the manner described above, and the y-axis is the probability density function p b () of the blur index within a range of 0.22 <
'No blur' on the top right of the graph is the unblurred natural image, and 'sigma' can be the standard deviation value of the Gaussian blur filter.
The higher the standard deviation value of the Gaussian blur filter, the stronger the blurring intensity with respect to the image.
In the case of 'No blur', which is a natural image with no manipulation, the maximum value of the graph can be close to 0 and it can have sharp tap value.
The comparison of the statistical parameters before and after blurring can be as shown in Table 1.
The scale parameter alpha may have a value of 1 or less in the case of a natural image to which no operation is applied. On the other hand, the scale parameter? Of the blurred image may all have a value of 1 or more.
The shape parameter c may be substantially higher than the blurred image in the case of a natural image to which no manipulation is applied.
The shape parameter k may have a value substantially smaller than that of a blurred image in the case of a natural image to which no manipulation is applied. However, the shape parameter k may be slightly different in the change tendency from image to image.
Referring again to FIG. 2, the statistical parameters estimated in step S150 are clustered (S160). For example, the counterfeit
In some embodiments, the clustering of statistical parameters may utilize a " k-means algorithm ".
Next, using the result of clustering the statistical parameters, a region where the statistical parameters are locally dense is determined as a fake area (S170). For example, the falsification
Hereinafter, effects according to the embodiments of the present invention will be described with reference to Figs. 9A and 9B.
FIGS. 9A and 9B are diagrams for explaining a result of detection of a false image according to the embodiments of the present invention.
Referring to FIG. 9A, (a) is an image to be identified, and (b) is a grouping of statistical parameters of the image to be checked according to embodiments of the present invention.
The confirmation target image (a) is obtained by inserting a monkey image into the original image and blurring the edge of the inserted monkey image.
(b), the statistical parameters of the image (a) to be confirmed are locally concentrated in the yellow and blue regions.
In this case, the fake
According to the forgery detection method according to the technical idea of the present invention, even if the synthesized monkey image itself has a natural blur (yellow area of (b)), the falsification area can be accurately determined.
Referring to FIG. 9B, (a) is an image to be checked, (b) is a result of judging a fake area of the image to be confirmed by a conventional method, and (c) The statistical parameters of the target image are clustered.
The confirmation target image (a) is obtained by synthesizing a face image of a specific entertainer on the body of the original image, and blurring the edge of the synthesized face image of the specific entertainer.
(c), the statistical parameters of the image to be confirmed (a) are locally concentrated in the green region.
In this case, the fake
In the case of the confirmation target image a, it is impossible to judge which part is a fake area in the case of using a conventional method as shown in (b) with a very precisely falsified image. However, as shown in (c), according to the falsification area detection method according to the technical idea of the present invention, the falsification area can be accurately determined.
That is, according to the falsification area detection method according to the technical idea of the present invention, the blur index is measured for all the edge points of the check target image, so that the blur index of the background area of the check target image that is not blurred is not detected, A more accurate forgery area can be judged.
The falsification area detecting method according to the technical idea of the present invention may be implemented in a general-purpose digital computer that can be created as a program that can be executed by a computer and operates the program using a computer-readable recording medium.
Hereinafter, with reference to FIG. 10, a computing system including a program for detecting false images according to embodiments of the present invention will be described.
10 is a diagram illustrating a computing system including a program for detecting false images in accordance with embodiments of the present invention.
The
The
The
The
The
The
For example, the
The counterfeit
Further, the counterfeit
Also, the counterfeit
Further, the counterfeit
Further, the counterfeit
The user can implement the falsification area detecting method according to the technical idea of the present invention in a form that downloads, stores or executes the falsification
While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, It is to be understood that the invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive.
Claims (8)
A blur estimator measures a blur index of an edge point of the confirmation target image detected by the edge detection unit,
A statistical parameter detector modeling a blur exponential function using the blur index measured by the blur index measuring unit, estimating a statistical parameter based on the modeled blur exponent function,
Wherein the forged region determinator clusters the statistical parameters estimated by the statistical parameter detector and determines a forgery region of the verification target image using the clustered statistical parameters,
The edge detecting unit detects an edge point of the check target image,
Measuring an edge sample at an edge point of the image to be identified,
The blur index,
And a cost function using the edge sample is minimized.
The falsification area determination unit clusters the statistical parameters estimated by the statistical parameter detection unit and determines the falsification area of the verification target image using the group,
And determining a locally dense portion of the statistical parameter as a fake area of the verification object image.
Wherein the edge point of the check target image includes a plurality of edge points,
The blur index measuring unit measures the blur index,
And measuring the blur index for the plurality of edge points.
A blur index of an edge point of the detected image to be checked is measured,
Modeling the blur exponential function using the measured blur exponent,
Estimating a statistical parameter based on the modeled Blur exponential function,
Clustering the estimated statistical parameters, determining a fake area of the verification target image using the clustered statistical parameters,
The detection of the edge point of the image to be confirmed,
Measuring an edge sample at an edge point of the image to be identified,
The blur index,
Wherein the edge function is a value that minimizes a cost function using the edge sample.
The determination of the forgery area of the image to be confirmed,
And determining a locally dense portion of the statistical parameter as a fake area of the verification object image.
Wherein the edge point of the check target image includes a plurality of edge points,
The blur index of the edge point of the detected image to be checked is measured,
And measuring the blur index for the plurality of edge points.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020150104105A KR101725041B1 (en) | 2015-07-23 | 2015-07-23 | Method and apparatus for detecting forged image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020150104105A KR101725041B1 (en) | 2015-07-23 | 2015-07-23 | Method and apparatus for detecting forged image |
Publications (2)
Publication Number | Publication Date |
---|---|
KR20170011454A KR20170011454A (en) | 2017-02-02 |
KR101725041B1 true KR101725041B1 (en) | 2017-04-10 |
Family
ID=58154072
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
KR1020150104105A KR101725041B1 (en) | 2015-07-23 | 2015-07-23 | Method and apparatus for detecting forged image |
Country Status (1)
Country | Link |
---|---|
KR (1) | KR101725041B1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20210084150A (en) | 2019-12-27 | 2021-07-07 | 한국항공대학교산학협력단 | System and method for the detection of multiple compression of image and video |
KR20220084236A (en) | 2020-12-13 | 2022-06-21 | 한국항공대학교산학협력단 | Advanced system and method for detecting video forgery |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110874845B (en) * | 2018-09-03 | 2022-06-21 | 中国科学院深圳先进技术研究院 | Method and device for detecting image smoothing |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101436406B1 (en) | 2011-12-29 | 2014-09-02 | 주식회사 안랩 | Client, server, system and method for updating data based on peer to peer |
-
2015
- 2015-07-23 KR KR1020150104105A patent/KR101725041B1/en active IP Right Grant
Non-Patent Citations (2)
Title |
---|
G Cao 외 2인, "Edge-based blur metric for tamper detection", Journal of Information Hiding and Multimedia Signal Processing, January 2010, vol 1, 20~27 pages* |
Matthew C. Stamm 외 1인, "Forensic detection of image manipulation using statistical intrinsic fingerprints", IEEE Transactions on Information Forensics and Security, 2010, Vol 5, Issue 3, pp. 492~506* |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20210084150A (en) | 2019-12-27 | 2021-07-07 | 한국항공대학교산학협력단 | System and method for the detection of multiple compression of image and video |
KR20220084236A (en) | 2020-12-13 | 2022-06-21 | 한국항공대학교산학협력단 | Advanced system and method for detecting video forgery |
Also Published As
Publication number | Publication date |
---|---|
KR20170011454A (en) | 2017-02-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109076198B (en) | Video-based object tracking occlusion detection system, method and equipment | |
KR101899866B1 (en) | Apparatus and method for detecting error of lesion contour, apparatus and method for correcting error of lesion contour and, apparatus for insecting error of lesion contour | |
US9269155B2 (en) | Region growing method for depth map/color image | |
JP6482195B2 (en) | Image recognition apparatus, image recognition method, and program | |
US8483480B2 (en) | Method and system for factoring an illumination image | |
TWI687689B (en) | Measurement device and measurement method for rotation of round body and non-transitory information readable medium | |
US9396411B2 (en) | Method and system for generating intrinsic images using a single reflectance technique | |
CN107292269B (en) | Face image false distinguishing method based on perspective distortion characteristic, storage and processing equipment | |
KR101761586B1 (en) | Method for detecting borderline between iris and sclera | |
US20140050411A1 (en) | Apparatus and method for generating image feature data | |
KR101725041B1 (en) | Method and apparatus for detecting forged image | |
TWI500904B (en) | Stereo camera and automatic range finding method for measuring a distance between stereo camera and reference plane | |
JP2015197376A (en) | Device, method, and program for abrasion detection | |
KR101662407B1 (en) | Method for vignetting correction of image and apparatus therefor | |
US9754155B2 (en) | Method and system for generating intrinsic images using a single reflectance technique | |
EP2791865B1 (en) | System and method for estimating target size | |
CN106886796B (en) | Icon position identification method and device and terminal equipment | |
JP4685711B2 (en) | Image processing method, apparatus and program | |
CN110889817B (en) | Image fusion quality evaluation method and device | |
JP6855938B2 (en) | Distance measuring device, distance measuring method and distance measuring program | |
CN114581433A (en) | Method and system for obtaining metal ball cavity inner surface appearance detection image | |
CN111951254A (en) | Source camera identification method and system based on edge-guided weighted average | |
JP4560434B2 (en) | Change region extraction method and program of the method | |
CN115019069A (en) | Template matching method, template matching device and storage medium | |
JP2009151445A (en) | Subarea detection device, object identification apparatus, and program |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
A201 | Request for examination | ||
E701 | Decision to grant or registration of patent right | ||
GRNT | Written decision to grant |