KR101871065B1 - 3-Dimensional Mesh Model watermarking Method Using Segmentation and Apparatus Therefor - Google Patents

Abstract

A method and apparatus for three-dimensional mesh model watermarking using segmentation techniques are disclosed. According to an embodiment of the present invention, there is provided a method of inserting a three-dimensional mesh model watermark, the method including: calculating a geometric feature value of a three-dimensional mesh model; Dividing the 3D mesh model into a plurality of mesh fragments using the geometric feature values; Converting the plurality of mesh fragments into a coordinate system using an apex distance; Inserting a predetermined watermark into each of a plurality of mesh fragments converted into a coordinate system using the vertex distance; And reconstructing the plurality of mesh fragments into which the watermark is embedded, into an orthogonal coordinate system.

Description

{3-Dimensional Mesh Model Watermarking Method Using Segmentation and Apparatus Therefor}

The present invention relates to a three-dimensional (3D) mesh model watermarking technique, and more particularly, to a three-dimensional (3D) mesh model watermarking technique that divides a 3D mesh model using a segmentation technique, or inserts or detects a watermark A mesh model watermarking method and apparatus.

As the 3D printer market expands and 3D model editing tools evolve, sharing of 3D models over the Internet becomes more frequent. However, the 3D model is a digital work, and because of its nature, it is easy to store and convert information, so that it can be reproduced without limitation and illegal distribution problem occurs. Therefore, there is a need for a method for protecting the copyright of the 3D model.

Traditional encryption methods can only protect digital works until decryption and once data is decrypted, data protection is impossible.

Digital watermarking is a copyright protection technology to overcome the problem of encryption method. Digital watermarking proves copyright by inserting and detecting specific data patterns in a digital work.

There are many digital watermarking methods to protect the copyright of the 3D model. However, since the existing methods are the method of inserting the watermark into the entire area of the 3D model, when the 3D model is divided, cut and pasted, There is a problem that watermark detection can not be performed.

In addition, since the process of splitting, cutting, and pasting a 3D model can easily occur, there is a need for a 3D model watermarking method that can protect the copyright of such a 3D model.

Embodiments of the present invention provide a watermarking method and apparatus capable of dividing a three-dimensional mesh model, inserting and detecting a watermark, and the like.

Embodiments of the present invention provide a watermarking method and apparatus capable of inserting and detecting a watermark that is robust against partial deformation such as splitting, cutting, and pasting a three-dimensional mesh model.

Embodiments of the present invention provide a watermarking method and apparatus capable of detecting a watermark without an original.

According to an embodiment of the present invention, there is provided a method of inserting a three-dimensional mesh model watermark, the method including: calculating a geometric feature value of a three-dimensional mesh model; Dividing the 3D mesh model into a plurality of mesh fragments using the geometric feature values; Converting the plurality of mesh fragments into a coordinate system using an apex distance; Inserting a predetermined watermark into each of a plurality of mesh fragments converted into a coordinate system using the vertex distance; And reconstructing the plurality of mesh fragments into which the watermark is embedded, into an orthogonal coordinate system.

The step of inserting the watermark includes: setting a watermark insertion period of each of the plurality of mesh fragments converted into a coordinate system using the vertex distance; And inserting the watermark into the set watermark insertion period of each of the plurality of mesh fragments converted into the coordinate system using the vertex distance.

The step of setting the watermark insertion interval may set the watermark insertion interval at a predetermined interval by histogramming each of the plurality of mesh fragments.

The dividing may divide the 3D mesh model into the plurality of mesh fragments using a clustering method after calculating the geometric feature values of the 3D mesh model.

According to another aspect of the present invention, there is provided a method of detecting a three-dimensional mesh model watermark, comprising: calculating geometric feature values of a three-dimensional mesh model; Dividing the 3D mesh model into a plurality of mesh fragments using the geometric feature values; Converting the plurality of mesh fragments into a coordinate system using an apex distance; Detecting a watermark from each of a plurality of mesh fragments converted into a coordinate system using the vertex distance; And determining a final watermark based on the detected watermark from each of the plurality of mesh fragments.

The step of detecting the watermark includes: setting a watermark detection period of each of the plurality of mesh fragments converted into a coordinate system using the vertex distance; And detecting the watermark in the set watermark detection period of each of the plurality of mesh fragments converted into the coordinate system using the vertex distance.

The step of setting the watermark detection interval may set the watermark detection interval at a predetermined interval by histogramming each of the plurality of mesh fragments.

The dividing may divide the 3D mesh model into the plurality of mesh fragments using a clustering method after calculating the geometric feature values of the 3D mesh model.

The determining of the final watermark may determine a final watermark from among the watermarks detected from each of the plurality of mesh fragments using a watermark trail analysis method or a majority voting system.

According to another aspect of the present invention, there is provided an apparatus for inserting a three-dimensional mesh model watermark, comprising: a calculation unit for calculating geometric feature values of a three-dimensional mesh model; A dividing unit dividing the 3D mesh model into a plurality of mesh fragments using the geometric feature values; A transform unit for transforming the plurality of mesh fragments into a coordinate system using an apex distance; An inserter for inserting a predetermined watermark into each of the plurality of mesh fragments converted into a coordinate system using the vertex distance; And a reconstruction unit that reconstructs the plurality of mesh fragments into which the watermark is embedded, into an orthogonal coordinate system.

Wherein the inserting unit sets a watermark inserting interval of each of the plurality of mesh fragments converted into the coordinate system using the apex distance and adds the watermark inserting interval of each of the plurality of mesh fragments converted into the spherical coordinate system to the set watermark inserting interval, Can be inserted.

The inserting unit may histogram each of the plurality of mesh fragments to set the watermark inserting interval at a predetermined interval.

The dividing unit may divide the 3D mesh model into the plurality of mesh fragments using a clustering method after calculating the geometric feature values of the 3D mesh model.

According to another aspect of the present invention, there is provided an apparatus for detecting three-dimensional mesh model watermark, comprising: a calculation unit for calculating geometric feature values of a three-dimensional mesh model; A dividing unit dividing the 3D mesh model into a plurality of mesh fragments using the geometric feature values; A transform unit for transforming the plurality of mesh fragments into a coordinate system using an apex distance; A detector for detecting a watermark from each of a plurality of mesh fragments converted into a coordinate system using the vertex distance; And a determination unit that determines a final watermark based on the watermark detected from each of the plurality of mesh pieces.

Wherein the detecting unit sets a watermark detection period of each of the plurality of mesh fragments converted into the coordinate system using the vertex distance and detects the watermark detection period of each of the plurality of mesh fragments converted into the coordinate system using the vertex distance The watermark can be detected.

The detection unit may histogram each of the plurality of mesh fragments to set the watermark detection interval at a predetermined interval.

The dividing unit may divide the 3D mesh model into the plurality of mesh fragments using a clustering method after calculating the geometric feature values of the 3D mesh model.

The determining unit may determine a final watermark among watermarks detected from each of the plurality of mesh fragments using a watermark tracing method or a majority voting system.

According to embodiments of the present invention, it is possible to insert and detect a watermark by dividing a three-dimensional mesh model, specifically, to embed a watermark that is robust against a partial transformation such as splitting, cutting, and pasting a three- Can be detected.

Specifically, embodiments of the present invention are robust to typical geometric transformations such as movement, scaling, and rotation generated in a three-dimensional mesh model, robust to connection information transformation such as mesh ordering, It is possible to exhibit a good performance even in the case of a shift of a center point movement of a paste or the like.

According to the embodiments of the present invention, a watermark can be detected without an original, and the copyright of the three-dimensional drawing can be effectively protected.

FIG. 1 is a flowchart illustrating a 3D mesh model watermark embedding method according to an embodiment of the present invention.
FIG. 2 illustrates an example of calculating a geometric diameter value of a triangular surface of a three-dimensional mesh model.
FIG. 3 shows an example of an algorithm for inserting the first character in the watermark string and an algorithm for inserting the second character.
4 is a flowchart illustrating a 3D mesh model watermark detection method according to an exemplary embodiment of the present invention.
FIG. 5 illustrates a 3D mesh model watermark embedding apparatus according to an embodiment of the present invention.
6 is a block diagram of a 3D mesh model watermark embedding apparatus according to an embodiment of the present invention.

Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings. However, the present invention is not limited to or limited by the embodiments. In addition, the same reference numerals shown in the drawings denote the same members.

Embodiments of the present invention can detect and insert a watermark strong in partial deformation such as dividing, cutting, and pasting a three-dimensional mesh model by dividing the three-dimensional mesh model and inserting and detecting the watermark That is the point.

3D mesh models are used in various industries including computer-aided design and 3D printers. The 3D mesh model is generated by a combination of components such as vertex, edge, face, polygon, and surface.

The 3D mesh model can be generated through file conversion in a three-dimensional voxel model.

FIG. 1 is a flowchart illustrating a 3D mesh model watermark embedding method according to an embodiment of the present invention.

Referring to FIG. 1, a watermark embedding method according to an embodiment of the present invention calculates a geometric feature value of a 3D mesh model, and divides the 3D mesh model into a plurality of mesh fragments using the calculated geometric feature values S110, S120).

Here, in step S110, the geometric diameter value of the triangular surface of the 3D mesh model can be used as the geometric feature value, and the geometric diameter value is the same as the example shown in FIG.

The geometric diameter value refers to the distance from the triangular plane to the point where the other triangular plane starts to draw a straight line inward of the mesh model. In order to calculate the more accurate geometric diameter value, it is possible to calculate the geometric diameter value by a certain number of angles while changing the angle within the allowable range in the form of a cone which is vertical and the inner direction of the mesh model. The representative geometric diameter value can be calculated by averaging the geometric diameter values. The green line shown in Fig. 2 means a geometric diameter line within an allowable angle, and the red line means a geometric diameter line deviating from the allowable angle, and the geometric diameter line can be normalized to a certain range such as [0, 1].

In step S120, the geometric feature values are calculated and then divided into a plurality of mesh fragments using a clustering method.

At this time, among the clustering methods, a Gaussian Mixture Model can be used. The Gaussian mixture model is defined as a linear combination of G probability density functions. As the G increases, the mesh model can be divided into smaller sizes. The probability density function of the Gaussian mixture model can be expressed as Equation have.

[Equation 1]

는 혼합 모델의 기본 성분을 이루는 확률밀도함수를 의미하고,

는 i번째 성분이 되는 확률밀도함수를 정의하는 파라미터로, 가우시안 확률밀도함수의 경우는 평균과 공분산 행렬이 될 수 있으며,

는 i번째 성분임을 나타내는 확률변수를 의미하고,

는 i번째 성분이 전체 혼합 확률밀도함수에서 차지하는 상대적인 중요도를 의미하며, 아래 <수학식 2>와 같은 성질을 만족해야 한다.here

Is a probability density function that forms the basic component of the mixed model,

Is a parameter that defines a probability density function that becomes an i-th component. In the case of a Gaussian probability density function, it may be an average and a covariance matrix.

Denotes a random variable indicating that it is an i-th component,

Denotes the relative importance of the i-th component in the entire mixed probability density function, and should satisfy the following Equation (2).

&Quot; (2) &quot;

Furthermore, in the present invention, in order to divide the 3D mesh model into a plurality of mesh fragments, a geometric primitive that can be calculated for each point or plane of the mesh can be used.

The geometric feature values that can be used in step S110 include various types of planarity, degree of similarity with the main figure such as sphere, cylinder, cone, degree of change of the back angle or normal vector, symmetry about curvature, slippage, geodesic distances, inter-mesh symmetry, convexity, center axis, and motion characteristics of the mesh animation.

In step S120, the 3D mesh model is divided into a plurality of mesh fragments using the computed geometric feature values. The 3D mesh model is classified into a region growing algorithm, a hierarchical clustering algorithm, an iterative clustering algorithm, A spectral analysis algorithm, a boundary setting algorithm, a top-down method, an induction algorithm, and the like can be used.

Region growing begins with a number of seeds and expands the region, including surrounding points with similar feature values.

Hierarchical clustering is a way to combine clusters with similar feature values.

Iterative clustering is a method that first establishes a model of cluster characteristics of feature values, 1) groups clusters according to the model, and 2) modifies the model according to cluster results. Repeating is the correct way to get the cluster model and results.

Spectral analysis is a method of transforming a mesh into a matrix and then dividing it by several mathematical operations.

The boundary setting is a method of indirectly dividing each part by setting the boundary of the division first.

The top-down method is a way to break up clusters into distinctly different pieces, as opposed to hierarchical clustering.

Induction is a method of calculating the skeleton of a mesh first and then dividing each part using it.

When the 3D mesh model is divided into a plurality of mesh fragments in step S120, each of the plurality of divided mesh fragments is converted into a coordinate system using an apex distance, and a plurality of mesh fragments A watermark insertion section for inserting a watermark is set (S130, S140).

The vertex distance refers to the distance from the origin to the vertex, and a spherical coordinate system and a cylindrical coordinate system can be used for the coordinate system using the vertex distance. Further, in the present invention, all the coordinate systems including the apex distance may be used as the components of the coordinate system.

로 나타낼 수 있으며, r은 원점에서부터 정점까지의 거리,

는 Z축의 양의 방향으로부터 원점과 정점이 이루는 직선까지의 각,

는 X축의 양의 방향으로부터 원점과 정점이 이루는 직선을 XY평면에 투영시킨 직선까지의 각도를 의미한다.The spherical coordinate system is

Where r is the distance from the origin to the vertex,

Is an angle from the positive direction of the Z axis to a straight line formed by the origin and apex,

Means an angle up to a straight line projecting a straight line formed by the origin and apex from the positive direction of the X axis on the XY plane.

At this time, for example, a plurality of L mesh pieces may be transformed from the orthogonal coordinate system to the spherical coordinate system through Equation (3) below.

&Quot; (3) &quot;

Here, (x _{l, g,} y _{l, g,} z _{l, g)} indicates the center of gravity of l second mesh piece, and (x _{l, n,} y _{l, n,} z _{l, n)} is the l-th mesh May denote the nth vertex coordinate of the piece.

The cylindrical coordinate system can be expressed as (ρ, φ, z ), where ρ is the distance from the origin to the vertex, φ is The angle from the positive direction of the X axis to the straight line projected from the origin to the vertex on the XY plane, and z means the distance from the XY plane to the vertex in the z axis direction.

In step S140, a plurality of, for example, L pieces of mesh pieces may be histogrammed to set a watermark insertion interval at a predetermined interval.

At this time, in step S140, the r of each of the mesh fragments may be histogrammed, and the histogram may be divided into a predetermined number of intervals, for example, M constant intervals longer than the length of the watermark string to be inserted. The range of r values can be normalized to a certain range such as [0, 1] or [-1, 1]. A watermark character string is inserted only in a predetermined interval of M consecutive intervals, and this interval is a watermark insertion interval.

Here, the watermark string may be a bit, an arbitrary numerical value, a character, or the like.

If a watermark insertion interval is set for each of the plurality of mesh fragments in step S140, a watermark is inserted in a watermark insertion interval set for each of the plurality of mesh fragments, and a plurality of mesh fragments, (S150, S160).

At this time, step S150 may insert one character of the watermark string to be inserted into each of the watermark insertion intervals set in step S140, and the character string uses two kinds of characters. The algorithm for inserting the first character of the string type may be as shown in FIG. 3A, and the algorithm for inserting the second character may be as shown in FIG. 3B.

는 정점을 의미하고, k는 정점을 변화시키는 파라미터로 한 번의 루프에 어느 정도 정점을 이동할지 결정하는 파라미터를 의미한다. k가 증가할수록 삽입 속도는 증가하지만 비가시성은 떨어질 수 있다.

는 워터마크 삽입 세기를 의미하고, 워터마크의 삽입이 완료되었을 때 어느 정도 정점을 이동할지 결정할 수 있다.

가 증가할수록 강인성은 증가하지만 비가시성은 떨어질 수 있다. V_{l ,m}는 l번째 메쉬 조각의 m번째 구간을 의미할 수 있다.3

Denotes a vertex, and k denotes a parameter for changing a vertex to determine a certain degree of vertex movement in one loop. As k increases, the insertion speed increases but the invisibility may decrease.

Means the watermark embedding strength, and it is possible to determine to what extent the apex is moved when the embedding of the watermark is completed.

, The toughness is increased but the invisibility can be lowered. V _{l , m} may mean the m-th section of the 1 st mesh piece.

The same watermark string is inserted into each of the plurality of mesh pieces through the process of step S150.

Furthermore, in the present invention, the interval statistic values such as mean, variance, standard deviation, slope, covariance, and correlation coefficient of each interval can be used as the watermark insertion method of step S150.

If the range of r values of each interval is normalized, the interval statistic value has a constant value. For example, if the range of r values in each interval is normalized to [-1, 1], the variance becomes close to 1/3, and the watermark can be inserted by increasing or decreasing the variance.

만큼 이동한다.By increasing the statistics of each interval, the first character of the watermark string type can be inserted, and the second character of the watermark string type can be inserted by lowering the interval statistic value. When inserting a watermark, the interval statistic value is the watermark insertion strength

.

If a watermark is inserted into each of the plurality of mesh fragments in step S150, the mesh fragments are restored to the orthogonal coordinate system (x, y, z) in step S160. If the coordinate system of the watermarked mesh fragments is a spherical coordinate system, the mesh fragments of the spherical coordinate system can be restored to the orthogonal coordinate system (x, y, z) using Equation (4) below.

&Quot; (4) &quot;

Here, (x ' _{l , n} , y' _{l , n} , z ' _{l , n} ) may mean the n th vertex coordinate of the l th mesh piece into which the watermark is inserted. r 'is the distance between the vertex and the origin where the watermark is inserted, and r is the distance between the vertex and the origin in the original where the watermark is not inserted.

As described above, in the watermark embedding method according to the embodiment of the present invention, a 3D mesh model is divided into a plurality of mesh fragments, and then converted into a coordinate system using an apex distance to insert a watermark string into each of the plurality of mesh fragments , It is robust against partial deformation such as splitting, cutting, and pasting of the 3D mesh model.

4 is a flowchart illustrating a 3D mesh model watermark detection method according to an exemplary embodiment of the present invention.

Referring to FIG. 4, a watermark detection method according to an exemplary embodiment of the present invention calculates a geometric feature value of a 3D mesh model to be watermarked, calculates a geometric feature value of the 3D mesh model, Into pieces (S410, S420).

Here, in step S410, the geometric feature value of the triangular surface of the 3D mesh model may be used as the geometric feature value. In step S420, the geometric feature value is calculated, and then, in order to divide the geometric feature value into a plurality of mesh pieces, (Gaussian Mixture Model).

Further, in the present invention, in order to divide the 3D mesh model into a plurality of mesh fragments, a geometric primitive that can be calculated for each point or each face of the mesh can be used.

The geometric feature values that can be used in step S410 include various types of planarity, degree of similarity with the main figure such as sphere, cylinder, cone, degree of change of the back angle or normal vector, symmetry about curvature, slippage, geodesic distances, inter-mesh symmetry, convexity, center axis, and motion characteristics of the mesh animation.

In step S420, the 3D mesh model is divided into a plurality of mesh fragments using the computed geometric feature values, and a region growing algorithm, a hierarchical clustering algorithm, an iterative clustering algorithm, A spectral analysis algorithm, a boundary setting algorithm, a top-down method, an induction algorithm, and the like can be used.

When the 3D mesh model is divided into a plurality of mesh fragments by step S420, each of the plurality of divided mesh fragments is converted into a coordinate system using the vertex distance, and a plurality of mesh fragments A watermark detection period for detecting a watermark is set (S430, S440).

At this time, in step S440, each of the plurality of mesh fragments may be histogrammed to set a watermark detection interval at a predetermined interval. The watermark detection period is the same as the preset watermark insertion period, and each detection period can be normalized.

If a watermark detection interval for each of the plurality of mesh fragments is set in step S440, a watermark is detected in the watermark insertion interval set for each of the plurality of mesh fragments, and the watermark detected from each of the plurality of mesh fragments (S450, S460).

Here, in step S450, the average value is calculated for each of the watermark detection periods set for the plurality of mesh fragments, and the first character of the watermark character string type is detected when the average value is 0.5 or more. If the average value is less than 0.5, The second character can be detected. That is, step S450 detects the watermark character string from the watermark detection interval set for each of the divided mesh fragments.

Further, in the present invention, when a statistical value is used for watermark insertion among the interval statistical values such as mean, variance, standard deviation, slope, covariance, and correlation coefficient of each section, the corresponding statistical value is used for watermark detection. For example, when using the variance as the interval statistics value when inserting a watermark and normalizing the range of r values of each interval as [-1, 1], the variance of each interval when the watermark is detected is higher than 1/3 If it is low, a watermark can be detected.

When the statistical value of each interval at the time of watermark detection is higher than the statistical value at the time of inserting the watermark, the first character of the watermark character string type can be detected. If the statistic value is lower than the statistical value at the time of inserting the watermark, Th character can be detected.

Step S460 is a process of determining a final watermark embedded in the 3D mesh model in the watermark detected in step S450, and may use a majority voting system (Watermark Trace Analysis) or a majority voting system.

Among them, the majority watermark system (Majority Voting System) can be used to determine the final watermark embedded in the 3D mesh model in the watermark detected in step S450. Equation (5) below is a mathematical expression of the majority voting system.

&Quot; (5) &quot;

은 l번째 메쉬 조각에서 검출한 워터마크 문자열을 의미하고, e₄는 결정된 최종 워터마크 문자열을 의미하고, BER(Bit Error Rate)은 두 문자열의 각 문자(또는 비트)를 비교한 계산값 예를 들어, 오류 비트수/전체 비트수를 의미할 수 있다.here,

E ₄ denotes the determined final watermark character string, and BER (Bit Error Rate) denotes an example of a calculation value in which each character (or bit) of two strings is compared with each other For example, it can mean the number of error bits / total number of bits.

Detecting the final watermark using the watermark detected in each mesh fragment can be formulated as Equation (6) below

&Quot; (6) &quot;

As can be seen from Equation (6), step S460 finds the most probable one as the original watermark when the watermark of each mesh fragment is given, and step S460 finds the most probable one among various methods, for example, Using a representative value such as an average of watermarks detected in each mesh slice, a machine learning method of setting and determining a probability model, and all probabilistic and statistical estimation methods, The mark can be determined.

The final watermark embedded in the 3D mesh model can be determined from the watermark detected in step S450 by using the watermark trace analysis method.

After the r of each mesh slice is histogrammed, the section corresponding to the watermark embedding interval is estimated by using a nonlinear least squares fitting method as shown in Equation (7) below.

&Quot; (7) &quot;

Where X is the set of vertices belonging to each section and Y is the actual value of X. The initial value of k for each interval in which no watermark is inserted is 1, and when the difference between the k value of the estimated interval function and the initial value of k is equal to or greater than a predetermined threshold value, it is determined as a watermark trail interval. Calculate the sum of the watermark trail sections for each 3D mesh segment. And determines the watermark of the 3D mesh segment whose sum of watermark trail sections is the maximum to be the final watermark.

Furthermore, in the present invention, statistical values such as mean, variance, standard deviation, slope, covariance, and correlation coefficient of each section may be used for each 3D mesh segment as a watermark trail section determination criterion. When the difference between the initial value of the statistical value before insertion of the watermark and the statistical value after insertion is equal to or greater than a predetermined threshold value, it is determined as a watermark trail section. The sum of the watermark trail sections is calculated for each 3D mesh piece and the watermark of the 3D mesh piece whose sum of the watermark trail sections is the maximum is determined as the final watermark.

As described above, in the 3D mesh model watermark detection method according to the embodiment of the present invention, in order to detect a watermark embedded in a plurality of mesh fragments, the 3D mesh model is divided into a plurality of mesh fragments, Coordinate system, and detects a watermark string embedded in each of the plurality of mesh segments, thereby determining a final watermark. Thus, it is robust against typical geometric transformation such as movement, scaling, and rotation generated in a 3D mesh model, It is also robust against connection information conversion such as sorting, and it is possible to easily detect a watermark embedded in center shift change such as division, cutting, pasting, and the like.

FIG. 5 is a block diagram of a 3D mesh model watermark embedding apparatus according to an embodiment of the present invention. FIG. 5 illustrates a structure of an apparatus for performing the watermark embedding method.

5, a 3D mesh model watermark embedding apparatus 500 according to an embodiment of the present invention includes a calculating unit 510, a dividing unit 520, a converting unit 530, an inserting unit 540, (550).

The calculation unit 510 calculates a geometric feature value of the 3D mesh model for division.

At this time, the calculation unit 510 can calculate the geometric diameter value of the triangular surface of the 3D mesh model to be used as the geometric feature value.

In this case, in order to divide the 3D mesh model into a plurality of mesh fragments, the calculation unit 510 may use a geometric primitive that can be calculated for each point or plane of the mesh, Planarity of a shape, degree of similarity with a main figure such as a sphere, cylinder, cone, degree of change of a backward angle or normal vector, curvature, slippage of rotation or position movement, geodesic distances, Mesh symmetry, convexity, center axis, and motion characteristics of the mesh animation.

The segmentation unit 520 divides the 3D mesh model into a plurality of mesh segments using the computed geometric feature values.

In this case, the dividing unit 520 may calculate the geometric feature values and then divide the mesh pieces into a plurality of mesh fragments using a clustering method.

At this time, the dividing unit 520 may perform a region growing algorithm, a hierarchical clustering algorithm, a repetitive clustering (see FIG. 4), and the like to divide the 3D mesh model into a plurality of mesh fragments using the calculated geometric feature values an iterative clustering algorithm, a spectral analysis algorithm, a boundary setting algorithm, a top-down method, and an inductive algorithm.

The transforming unit 530 transforms each of the plurality of divided mesh fragments into a coordinate system using an apex distance.

That is, the transforming unit 530 transforms each of the plurality of divided mesh fragments into a coordinate system using the vertex distance in the orthogonal coordinate system.

The inserting unit 540 inserts a watermark for each of the plurality of mesh fragments converted into the coordinate system using the vertex distance.

At this time, the inserting unit 540 sets a watermark inserting interval for inserting a watermark into each of the plurality of mesh fragments converted into the coordinate system using the vertex distance, and sets the watermark inserting interval for each of the plurality of mesh fragments The watermark is inserted in the insertion section.

The inserting unit 540 may set a watermark inserting interval of a predetermined interval by histogramming each of the divided mesh fragments and may insert one of the watermark strings to be inserted into each of the set watermark inserting intervals .

Accordingly, the inserting unit 540 inserts the same watermark character string into each of the plurality of mesh pieces.

The restoring unit 550 restores each of the plurality of mesh fragments into which the watermark is inserted from the coordinate system using the vertex distance to the orthogonal coordinate system.

FIG. 6 illustrates a structure of a 3D mesh model watermark embedding apparatus according to an embodiment of the present invention, and shows a structure of an apparatus for performing the watermark detection method.

6, a 3D mesh model watermark detection apparatus 600 according to an embodiment of the present invention includes a calculation unit 610, a division unit 620, a conversion unit 630, a detection unit 640, 650).

The calculation unit 610 calculates a geometric feature value of the 3D mesh model for division.

At this time, the calculation unit 610 can calculate the geometric diameter value of the triangular surface of the 3D mesh model to be used as the geometric feature value.

In this case, in order to divide the 3D mesh model into a plurality of mesh fragments, the calculation unit 610 may use a geometric primitive that can be calculated for each point or plane of the mesh, Planarity of a shape, degree of similarity with a main figure such as a sphere, cylinder, cone, degree of change of a backward angle or normal vector, curvature, slippage of rotation or position movement, geodesic distances, Mesh symmetry, convexity, center axis, and motion characteristics of the mesh animation.

The segmentation unit 620 divides the 3D mesh model into a plurality of mesh segments using the computed geometric feature values.

At this time, the dividing unit 620 may calculate the geometric feature values and then divide the mesh fragments into a plurality of mesh fragments using the clustering method.

In this case, the partitioning unit 620 may use a region growing algorithm, a hierarchical clustering algorithm, and a repetitive clustering (hereinafter referred to as &quot; clustering &quot;) algorithm to divide the 3D mesh model into a plurality of mesh fragments using the computed geometric feature values an iterative clustering algorithm, a spectral analysis algorithm, a boundary setting algorithm, a top-down method, and an inductive algorithm.

The transforming unit 630 transforms each of the plurality of divided mesh fragments into a coordinate system using an apex distance.

That is, the transforming unit 630 transforms each of the plurality of divided mesh fragments into a coordinate system using the vertex distance in the orthogonal coordinate system.

The detecting unit 640 detects a watermark for each of the plurality of mesh fragments converted into the coordinate system using the vertex distance.

At this time, the detection unit 640 sets a watermark detection period for detecting a watermark for each of a plurality of mesh fragments converted into a coordinate system using an apex distance, and detects a watermark detection for each of the plurality of mesh fragments The watermark is detected in the interval.

The detection unit 640 may set a watermark detection interval of a predetermined interval by histogramming each of the divided mesh fragments.

The determination unit 650 determines the final watermark based on the watermark detected from each of the plurality of mesh pieces.

At this time, the determining unit 650 calculates an average value for each of the watermark detection intervals set for the plurality of mesh fragments, detects the first character of the watermark string when the average value is 0.5 or more, and when the average value is less than 0.5, The second character of the mark string can be detected. That is, the determination unit detects the watermark character string from the watermark detection interval set for each of the divided mesh fragments.

At this time, the determination unit 650 determines specific statistical values among the interval statistical values such as average, variance, standard deviation, slope, covariance, and correlation coefficient of each interval for each of the watermark detection intervals set for a plurality of mesh fragments When used for watermark insertion, the statistical value is used for watermark detection. That is, when the statistical value of each section at the time of watermark detection is higher than the statistical value at the time of watermark embedding, the first character of the watermark character string type can be detected. When the statistical value is lower than the statistical value at the time of watermark embedding, The second character is detected.

The determination unit 650 determines the final watermark embedded in the 3D mesh model in the watermark detected in each of the plurality of mesh pieces.

At this time, the determination unit 650 can determine the final watermark embedded in the 3D mesh model in the watermark detected in each of the plurality of mesh fragments using a majority voting system (Majority Voting System).

At this time, the determination unit 650 can determine the final watermark embedded in the 3D mesh model in the watermark detected in each of the plurality of mesh fragments using Watermark Trace Analysis.

The system or apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the systems, devices, and components described in the embodiments may be implemented in various forms such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array ), A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to embodiments may be implemented in the form of a program instruction that may be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI &gt; or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (18)

Calculating geometric feature values of the three-dimensional mesh model;
Dividing the 3D mesh model into a plurality of mesh fragments using the geometric feature values;
Converting a center of gravity of the plurality of divided mesh pieces into a coordinate system using an apex distance indicating a distance from a center of gravity of each mesh piece to a vertex in the mesh piece;
Inserting a predetermined watermark into each of a plurality of mesh fragments converted into a coordinate system using the vertex distance; And
And reconstructing the plurality of mesh fragments into which the watermark has been inserted into an orthogonal coordinate system,
The step of inserting the watermark
Wherein the watermark embedding method uses interval statistical values such as mean, variance, standard deviation, slope, covariance, and correlation coefficient of each interval, and the interval statistic value moves by the watermark embedding intensity. Method of inserting mesh model watermark.
The method according to claim 1,
The step of inserting the watermark
Setting the watermark embedding period of each of the plurality of mesh fragments converted into a coordinate system using the vertex distance; And
Inserting the watermark into the set watermark insertion period of each of the plurality of mesh fragments converted into the coordinate system using the vertex distance
Dimensional mesh model watermark embedding method.
3. The method of claim 2,
The step of setting the watermark insertion section
Wherein each of the plurality of mesh fragments is histogrammed to set the watermark insertion interval at a predetermined interval.
The method according to claim 1,
The dividing step
Wherein the geometric feature values of the three-dimensional mesh model are calculated and then the three-dimensional mesh model is divided into the plurality of mesh pieces using a clustering method.
Calculating geometric feature values of the three-dimensional mesh model;
Dividing the 3D mesh model into a plurality of mesh fragments using the geometric feature values;
Converting a center of gravity of the plurality of divided mesh pieces into a coordinate system using an apex distance indicating a distance from a center of gravity of each mesh piece to a vertex in the mesh piece;
Detecting a watermark from each of a plurality of mesh fragments converted into a coordinate system using the vertex distance; And
Determining a final watermark based on the detected watermark from each of the plurality of mesh fragments,
The step of detecting the watermark
A watermark character string is detected from a watermark detection interval set for each of the plurality of mesh fragments, and a specific statistical value among interval statistical values such as average, variance, standard deviation, slope, covariance, and correlation coefficient of each interval is used for watermark insertion , The statistical value is used for watermark detection,
The step of determining the final watermark
When the difference between the initial value of the statistical value before insertion of the watermark and the statistical value after insertion is equal to or greater than a predefined threshold value, it is determined as a watermark trail section, and the sum of the watermark trail sections And the final watermark is determined by calculating the final watermark.
6. The method of claim 5,
The step of detecting the watermark
Setting a watermark detection interval of each of the plurality of mesh fragments converted into a coordinate system using the vertex distance; And
Detecting the watermark in the set watermark detection period of each of the plurality of mesh fragments converted into the coordinate system using the vertex distance
Dimensional mesh model watermark detection method.
The method according to claim 6,
The step of setting the watermark detection section
Wherein each of the plurality of mesh fragments is histogrammed to set the watermark detection interval at a predetermined interval.
6. The method of claim 5,
The dividing step
Calculating a geometric feature value of the three-dimensional mesh model, and then dividing the three-dimensional mesh model into the plurality of mesh pieces using a clustering method.
6. The method of claim 5,
The step of determining the final watermark
Wherein a final watermark is determined using a watermark trail analysis method or a majority vote system among watermarks detected from each of the plurality of mesh fragments.
A calculation unit for calculating a geometric feature value of the three-dimensional mesh model;
A dividing unit dividing the 3D mesh model into a plurality of mesh fragments using the geometric feature values;
A transformation unit for transforming the center of gravity of the divided plurality of mesh pieces into a coordinate system using a vertex distance representing a distance from a center of gravity of each mesh piece to a vertex in the mesh piece;
An inserter for inserting a predetermined watermark into each of the plurality of mesh fragments converted into a coordinate system using the vertex distance; And
And a reconstruction unit for reconstructing the plurality of mesh fragments into which the watermark is inserted into an orthogonal coordinate system,
The insert
Wherein the watermark embedding method uses interval statistical values such as mean, variance, standard deviation, slope, covariance, and correlation coefficient of each interval, and the interval statistic value moves by the watermark embedding intensity. Mesh model watermark embedding device.
11. The method of claim 10,
The insert
A step of setting the watermark inserting interval of each of the plurality of mesh fragments converted into the coordinate system using the vertex distance and setting the watermark inserting interval of each of the plurality of mesh fragments converted into the coordinate system using the vertex distance, Mark is inserted into the three-dimensional mesh model.
12. The method of claim 11,
The insert
Wherein each of the plurality of mesh fragments is histogrammed to set the watermark insertion interval at a predetermined interval.
11. The method of claim 10,
The divider
Wherein the geometric feature values of the three-dimensional mesh model are calculated and then the three-dimensional mesh model is divided into the plurality of mesh pieces using a clustering method.
A calculation unit for calculating a geometric feature value of the three-dimensional mesh model;
A dividing unit dividing the 3D mesh model into a plurality of mesh fragments using the geometric feature values;
A transformation unit for transforming the center of gravity of the divided plurality of mesh pieces into a coordinate system using a vertex distance representing a distance from a center of gravity of each mesh piece to a vertex in the mesh piece;
A detector for detecting a watermark from each of a plurality of mesh fragments converted into a coordinate system using the vertex distance; And
And a determination unit that determines a final watermark based on the detected watermark from each of the plurality of mesh fragments,
The detection unit
A watermark character string is detected from a watermark detection interval set for each of the plurality of mesh fragments, and a specific statistical value among interval statistical values such as average, variance, standard deviation, slope, covariance, and correlation coefficient of each interval is used for watermark insertion , The statistical value is used for watermark detection,
The determination unit
When the difference between the initial value of the statistical value before insertion of the watermark and the statistical value after insertion is equal to or greater than a predefined threshold value, it is determined as a watermark trail section, and the sum of the watermark trail sections And the final watermark is determined by calculating the final watermark.
15. The method of claim 14,
The detection unit
And a step of setting a watermark detection period of each of the plurality of mesh fragments converted into the coordinate system using the vertex distance and extracting the watermark detection period of each of the plurality of mesh fragments converted into the coordinate system using the vertex distance, Of the three-dimensional mesh model watermark detection apparatus.
16. The method of claim 15,
The detection unit
Wherein each of the plurality of mesh fragments is histogrammed to set the watermark detection interval at a predetermined interval.
15. The method of claim 14,
The divider
Wherein the 3D mesh model is divided into the plurality of mesh fragments using a clustering method after calculating the geometric feature values of the 3D mesh model.
15. The method of claim 14,
The determination unit
Wherein the final watermark is determined using a watermark tracing analysis method or a majority vote system among watermarks detected from each of the plurality of mesh fragments.

Images