KR20110088995A - Method and system to visualize surveillance camera videos within 3d models, and program recording medium - Google Patents

Abstract

PURPOSE: A method, a system and recording medium for visualizing a monitoring camera image inside a 3D model are provided to express a 2D screen or 3D space without distortion. CONSTITUTION: A virtual object realizing an object located in at least one monitoring camera as a virtual 3D model is generated(S310). An image of at least one monitoring camera is projected on a screen which faces at least one monitoring camera. For rendering, the image is adjusted to be vertical to a visible direction of a virtual camera which is virtually set the surface(S320).

Description

Method and System To Visualize Surveillance Camera Videos Within 3D Models, and Program Recording Medium}

The present invention relates to a method and system for visualizing a surveillance camera image, and more particularly, to a method and system for visualizing a surveillance camera image in a three-dimensional model.

Surveillance camera (CCTV) control software (e.g., Digital Video Recording S / W) screen shown in Figure 1 basically corresponds to a two-dimensional matrix (matrix), through this screen the user is looking at the three-dimensional camera It is not easy to know intuitively where it is in space.

Although there is a simple label for each surveillance camera screen 11-19, a considerable amount of time is required for the user to connect the camera screen 11-19 with three-dimensional or two-dimensional positional information directly in the imagination. need.

Therefore, there is a need for a new visualization method that allows a user to view a video image and intuitively determine the image in time and space, and furthermore, in a complex environment (e.g., a complex three-dimensional space if there are many surveillance cameras), What is needed is a visualization method that can provide a controllable environment.

Therefore, the technical problem to be achieved by the present invention is to provide a new visualization method that allows the user to intuitively determine the image in time and space by looking at the video image and a method capable of efficiently controlling the environment even in a complex environment. To provide.

According to an embodiment of the present invention, there is provided a method for visualizing an image on a three-dimensional model, including: generating a virtual object in which an object in which at least one surveillance camera is located is implemented as a virtual three-dimensional model; Projecting an image of the at least one surveillance camera onto a plane facing the at least one surveillance camera; And adjusting the plane to be perpendicular to a line of sight of a virtual camera, which is a camera virtually set for rendering.

In addition, the method for visualizing the image on the three-dimensional model is displayed on the virtual object, at least one view window of the preset size of the two-dimensional image taken by each of the surveillance cameras appearing in the adjusted plane It may include the step.

In addition, the method for visualizing the image on the three-dimensional model, when the user selects one of the at least one view window, projecting the image taken by the surveillance camera of the selected view window on the virtual object and outputs It may include a step.

The method for visualizing an image on the 3D model may include projecting and outputting an image photographed by the virtual camera and a surveillance camera close to a line of sight or a distance on the virtual object.

In addition, when a predetermined situation occurs in the at least one view window, a color constituting the at least one view window may be changed, or the at least one view window may flicker.

The displaying through the at least one view window may include determining a color value of the at least one view window in a predetermined order; And comparing the color value with a reference value and moving the at least one view window in a predetermined direction to display the color value.

In addition, displaying through the at least one view window may include displaying the view windows by connecting an annotation line with a corresponding surveillance camera position on the virtual object.

In addition, displaying through the at least one view window may include displaying a view window of surveillance cameras in the selection area when a specific area on the virtual object is selected.

The generating of the virtual object may include wireframe rendering and / or polygon rendering or alpha value adjusted rendering to visualize the surveillance region of the surveillance image and the surveillance camera. Generating and rendering the virtual object.

In addition, the polygon rendering may use a shadow-map algorithm.

The generating of the virtual object may include three-dimensional vectors in a specific direction directed to vanishing points of the captured image and an external vector based on the three-dimensional vectors based on the captured image of the at least one surveillance camera. Generating a three-dimensional coordinate system by using; And positioning the at least one surveillance camera in the virtual object after calculating the position and / or the surveillance direction of the at least one surveillance camera by positioning the three-dimensional coordinate system on the virtual object according to a predetermined criterion. It may include.

In addition, the system for visualizing the image on the three-dimensional model according to the present invention generates a virtual object that implements an object (at) the at least one surveillance camera is located in a virtual three-dimensional model, and of the at least one surveillance camera A processor programmed to project an image onto a plane facing the at least one surveillance camera and adjust the plane to be perpendicular to a line of sight of a virtual camera, a camera virtually set for rendering; And a database for storing data input and / or output to the system for processing by the processor.

The processor may generate a virtual object that implements an object in which at least one surveillance camera is located as a virtual three-dimensional model, and may display an image of the at least one surveillance camera in front of the at least one surveillance camera. A modeling module for projecting on a plane to be viewed and modeling the plane to be perpendicular to a direction of the eye of a virtual camera, which is a camera virtually set for rendering; A detection module for detecting and transmitting data input and / or output to the system for modeling the modeling module; And a color buffer for rendering the data modeled by the modeling module as color values.

The system for visualizing an image on the 3D model may include a display device for displaying a result processed by the processor.

According to the present invention, a two-dimensional screen can be represented without distortion in a three-dimensional space represented in two or three dimensions.

In addition, the present invention can provide a management system that is convenient for the user.

In addition, according to the present invention there is an effect that can be implemented to match the three-dimensional virtual space more realistic.

1 illustrates a screen output by conventional surveillance cameras.
2 illustrates a system for visualizing a surveillance camera image in a three-dimensional model according to an embodiment of the present invention.
3A and 3B are diagrams for describing a method for visualizing a surveillance camera image in a 3D model according to a comparative example of the present invention.
3C to 3E are diagrams for describing a method for visualizing a surveillance camera image in a 3D model according to an embodiment of the present invention.
3F is a schematic flow diagram illustrating a method for visualizing a surveillance camera image within a three-dimensional model of the present invention.
4A and 4B are diagrams illustrating an example in which a method for visualizing a surveillance camera image in a 3D model according to FIGS. 3C to 3F is implemented.
4C and 4D are diagrams for explaining an algorithm for indicating the position and surveillance direction of the surveillance camera displayed by the system of FIG.
5A is a diagram illustrating an example in which an image captured by a surveillance camera is projected and displayed on a 3D model.
FIG. 6A is a diagram for describing a method for visualizing a surveillance camera image in a 3D model when the number of surveillance cameras is large.
FIG. 6B is a diagram for one example of implementing a method for visualizing a surveillance camera image in a 3D model according to FIG. 6A.
FIG. 7A is a diagram illustrating a method for visualizing a surveillance camera image in a case where a 3D model is concentrated in a large area.
FIG. 7B is a diagram for illustrating a method for visualizing a surveillance camera image for an area where a 3D model is concentrated when the number of surveillance cameras is large.
FIG. 8 is a diagram illustrating a method for visualizing a surveillance camera image in a three-dimensional model in a plan view.
9A and 9B are diagrams for explaining an algorithm for detecting an accurate position and a surveillance direction of a surveillance camera on a three-dimensional model based on the captured image.

Specific structural to functional descriptions of the embodiments of the present invention disclosed in the specification or the application are only illustrated for the purpose of describing the embodiments according to the present invention, and the embodiments according to the present invention may be embodied in various forms. It should not be construed as limited to the embodiments described in this specification or the application.

Since the embodiments according to the present invention can be variously modified and have various forms, specific embodiments will be illustrated in the drawings and described in detail in the present specification or application. However, this is not intended to limit the embodiments in accordance with the concept of the present invention to a particular disclosed form, it should be understood to include all changes, equivalents, and substitutes included in the spirit and scope of the present invention.

Terms such as first and / or second may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another, for example, without departing from the scope of rights in accordance with the inventive concept, and the first component may be called a second component and similarly The second component may also be referred to as the first component.

When a component is referred to as being "connected" or "connected" to another component, it may be directly connected to or connected to that other component, but it may be understood that other components may be present in between. Should be. On the other hand, when a component is said to be "directly connected" or "directly connected" to another component, it should be understood that there is no other component in between. Other expressions describing the relationship between components, such as "between" and "immediately between," or "neighboring to," and "directly neighboring to" should be interpreted as well.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. As used herein, the terms "comprise" or "having" are intended to indicate that there is a feature, number, step, action, component, part, or combination thereof that is described, and that one or more other features or numbers are present. It should be understood that it does not exclude in advance the possibility of the presence or addition of steps, actions, components, parts or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art. Terms such as those defined in the commonly used dictionaries should be construed as having meanings consistent with the meanings in the context of the related art, and are not construed in ideal or excessively formal meanings unless expressly defined herein. Do not.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Like reference numerals in the drawings denote like elements.

2 illustrates a system for visualizing a surveillance camera image in a three-dimensional model according to an embodiment of the present invention. Referring to FIG. 2, the system 100 includes a processor 110 and a database 120.

The processor 110 may include a detection module 111, a modeling module 112, and an I / F 113, and are responsible for the overall processing of the system 110.

The detection module 111 may detect a position and a monitoring direction of the surveillance camera, detect all data related to a captured image received from the surveillance camera, and detect a pre-input situation when it occurs. .

The detection module 111 may detect data input or output to the system 100 for modeling the modeling module 112.

The modeling module 112 may configure a 3D model, and may perform 3D modeling according to a predetermined algorithm. The algorithm may include an algorithm for projecting a captured image of a surveillance camera onto a 3D model and a surveillance camera. Algorithm for optimally displaying the image of the user, and the like will be described in detail later.

In addition, the modeling module 112 may assign identification (ID) to predetermined data.

The I / F 113 is responsible for an interface between the detection module 111, the modeling module 112, and the database 120 to transmit a control command corresponding to the predetermined control by the user. do.

The database 120 may store all data input or output to the system 100 or data predetermined by a user.

The electronic device 200 may be connected to the system 100 through an I / F 113 and may correspond to a display device that displays information for visual or stereoscopic reception, storage, and transmission. For example, the electronic device 200 may include a computer monitor, a TV, a projector, and the like, but the scope of the present invention is not limited thereto.

In FIG. 2, the electronic device 200 is illustrated as being located outside the system 100, but may be located inside.

The processor 110 may further include a color buffer (not shown). The color buffer (not shown) may render predetermined data having an ID assigned by the modeling module 112 as a color value.

3A and 3B are diagrams for describing a method for visualizing a surveillance camera image in a 3D model according to a comparative example of the present invention. Here, the virtual camera is a camera that is distinguished from the surveillance camera and corresponds to a camera that is virtually set for rendering. In addition, the three-dimensional model can be navigated by rotating, zooming, panning and / or moving the virtual camera.

Referring to FIG. 3A, in order to visualize the surveillance camera image in the 3D model, the surveillance camera CAM0 may be displayed on the plane PN1 at a predetermined distance from the surveillance camera 31 in the shooting direction 31. CAM0) can be projected onto the three-dimensional model 30 (PN2). At this time, the field of view (FOV: Field of View) of the virtual camera VCAM0 should be matched with the field of view of the surveillance camera.

However, the methods may provide a distorted image to the user according to the position (or FOV) of the virtual camera VCAM0 as shown in FIG. 3B, and the methods may have difficulty in locating the surveillance camera CAM0. Can also give

3C to 3E are diagrams for describing a method for visualizing a surveillance camera image in a 3D model according to an embodiment of the present invention.

Referring to FIG. 3C, in order to visualize the surveillance camera image in the three-dimensional model 30, the surveillance camera CAM0 image is first viewed from the surveillance camera CAM0 while facing the surveillance camera CAM0. It can be shown in the plane (PN3) apart from. At this time, the gaze range of the virtual camera VCAM0 should match the gaze range of the surveillance camera.

In this case, the surveillance camera CAM0 may project a photographing screen on the plane PN3 along a viewing ray with respect to each pixel of the surveillance camera CAM0.

Adjusting the plane PN3 to be perpendicular to the eye direction S2 of the virtual camera VCAMO and the up direction 34 of the plane PN3 is parallel to the up direction 34 of the virtual camera VCAMO. The position PN3 'may be positioned at a predetermined distance 33 along the up direction 34 from the surveillance camera CAMO.

Therefore, as shown in FIG. 3D, the user can view the output image of the surveillance camera without distortion, regardless of the gaze of the virtual camera.

In this case, the projection may correspond to a case of projecting captured images of all surveillance cameras, and may also correspond to a case of projecting a surveillance camera near a line of sight or distance of the virtual camera.

The size of the output image may be proportional to the distance from the virtual camera VCAM0 to the surveillance camera CAM0. For example, when the distance from the virtual camera VCAM0 to the surveillance camera CAM0 is close, the size of the image appears larger, and when the distance from the virtual camera VCAM0 to the surveillance camera CAM0 is far, the size of the image appears smaller. Can be. Therefore, the user may be provided with an image that allows the user to feel a sense of depth according to the distance from the virtual camera VCAM0 to the surveillance camera CAM0.

In this case, by providing the surveillance camera photographed image to the user using the annotation line 32, it is possible to intuitively determine the position of the camera. As described above, the surveillance camera photographed image may be displayed to the user through a predetermined view window (eg, a rectangular frame for providing a camera image) connected to the annotation line 32 and positioned at a predetermined distance 33. Can be.

In addition, the size of the captured image of the surveillance camera may be adjusted. For example, the user may adjust the size of the surveillance camera photographed image through a predetermined function (eg, a button interface). When the size of the captured image of the surveillance camera is significantly larger than the size of the entire screen, the user may automatically switch to the matrix mode, which is a general control form, to provide 3D and / or 2D control to the user.

Referring to FIG. 3E, when viewed from the virtual camera VCAM0, the first surveillance camera CAM5 may be covered by the third surveillance camera CAM4 unlike the second surveillance camera CAM6. have.

Thus, an algorithm for taking this into account is described below.

First, the view windows of the surveillance cameras CAM4 to CAM6 are sequentially read and assigned a unique ID in a predetermined order, and the ID is set to a color value by using OpenGL FLAT shading of a color buffer. Render the captured image.

Next, the detection module 111 determines the number of pixels having the color value (or the ID value) in the view window. If the detection module 111 determines that the number of pixels is less than a predetermined number (ie, the view window is not well seen), the modeling module 112 moves the view window in a predetermined direction.

If the process is repeated and it is determined that the number of pixels is greater than or equal to a predetermined number (that is, the view window is judged to be well visible), the process ends.

The algorithm may be performed when the positions of the at least one surveillance cameras CAM4 to CAM6 change or the position or line of sight of the virtual camera VCAM0 changes.

3F is a schematic flow diagram illustrating a method for visualizing a surveillance camera image within a three-dimensional model of the present invention.

Referring to FIG. 3F, a virtual 3D model of a building or an area in which surveillance cameras are installed is configured (S310).

Next, the image of the surveillance camera is rendered on a plane away from the surveillance camera while looking at the surveillance camera in front (S320), and the captured image is output to the view window (S330). In this case, the surveillance camera may project a photographing screen on the plane according to the eye direction of each pixel of the surveillance camera CAM0. In addition, the view window is adjusted to be perpendicular to the eye direction of the virtual camera and the up direction of the view window is parallel to the up direction of the virtual camera, so that the view window may be positioned at a distance along the up direction from the surveillance camera. Can be.

Next, it is determined whether the view window is well seen (S340), if not, move the view window in a predetermined direction (S350) and repeat until it is determined that the view window is well seen.

4A and 4B are diagrams illustrating an example in which the method for visualizing a surveillance camera image in the 3D model according to FIGS. 3C to 3F is implemented by the system 100 of FIG. 2.

Referring to FIG. 4A, the image processed by the system 100 is a view window 41 of a predetermined size representing the 3D full image 40 and the 2D image monitored by the surveillance cameras A1 to A6. To 46). The view windows 41 to 46 are connected to the positions of the surveillance cameras A1 to A6 via tin lines 410 to 460.

Such a display enables the user to easily determine the position of the surveillance cameras A1 to A6, and more easily manage the system.

When the predetermined situation occurs, the view windows 41 to 46 may be modified so that the user may immediately detect the view windows 41 to 46. For example, the system 100 may cause the view windows 41 to 46 to blink or to change the color of the view windows 41 to 46. In this case, OpenGL color blending and / or programmable shaders may be used to change the color.

In addition, the system 100 may indicate a monitoring direction (not shown) that the surveillance cameras A1 to A6 are monitoring.

In addition, the user may adjust the size of the three-dimensional whole image 40 and / or the view windows 41 to 46 through a predetermined function (eg, a button interface) as shown in FIG. 4A, or The view windows 41 to 46 may be moved using an input device (eg, a mouse or a keyboard) connected to the system 100.

In addition, the view point and the direction of the current virtual camera, the position and / or size of the view windows 41 to 46, and the like may be stored.

The system 100 automatically converts the screen into a matrix mode when the screen image of the view windows 41 to 46 is larger than a predetermined size by the user, thereby allowing the user to three-dimensionally. And two-dimensional control may be possible.

4C and 4D are diagrams for explaining an algorithm for indicating the position and surveillance direction of the surveillance camera displayed by the system of FIG.

Referring to FIGS. 4C and 4D, when the positions and the monitoring directions 56 and 57 of the surveillance cameras CAM1 and CAM2 are positioned in the 3D image as shown in FIG. 4C, the surveillance camera CAM2 is moved by the 3D image. There may be a problem that the position and the monitoring direction 56 of the surveillance camera CAM2 are not accurately known.

Therefore, it may be necessary to display the surveillance directions of the surveillance cameras CAM1 and CAM2 and the surveillance cameras CAM1 and CAM2 in the three-dimensional model. That is, it may be necessary to display a field of view (FOV) of the surveillance cameras CAM1 and CAM2 and the surveillance camera itself on the screen.

As shown in FIG. 4D, only the planes captured by the surveillance cameras CAM1 and CAM2 are rendered in a polygon (51 to 53), and the rest are wireframe rendered, or the alpha value of the plane is adjusted. It is possible to solve the phenomenon in which the surveillance cameras CAM1 and CAM2 are covered by the 3D model by using an alpha value adjusting rendering method and the like. In this case, the surveillance cameras CAM1 and CAM2 may also render polygons (54 and 55).

The system 100 may perform the following algorithm to perform this.

First, rendering is performed by wireframe rendering or adjusting an alpha value of a plane for the entire 3D image. However, the present invention is not limited to the above-described rendering method when rendering the entire 3D.

Next, polygon rendering is performed using a shadow-map algorithm.

An example of a method of performing the polygon rendering using the shadow-map algorithm is described below.

First, a logical buffer object (eg, FBO: Framebuffer Object in OpenGL, Render Target in DirectX) is set and created at the position of the surveillance camera, and a depth texture is connected to the logical buffer object.

The model-view and projection matrix of the surveillance camera are stored. The three-dimensional model (object) is rendered at each surveillance camera position, and the rendered result is stored as a depth value in a shadow-map connected to the logical buffer object.

Rendering is done using a vertex / pixel shader at the virtual camera position. In this case, the vertex / pixel shader obtains coordinates based on the surveillance camera at each point inside each polygon by using the surveillance camera model-view and the projection matrix. The x and y axes are the texture coordinates of the shadow-map and the z axis is the depth value of the point.

Compare the polygon internal point depth value at the point with the depth value stored in the shadow-map and if it is the same, output this point to the screen, otherwise do not output it.

5A is a diagram illustrating an example in which an image captured by a surveillance camera is projected and displayed on a 3D model. When the user selects one view window (or when the virtual camera approaches a predetermined distance from the surveillance camera), the system 100 may display an image 72 provided by the selected view window. In addition, the image 71 captured by the surveillance camera associated with the selected view window may be projected on the 3D model. This allows the user to receive the same or similar feeling as observing the actual site.

FIG. 5B is a diagram for describing in more detail a method of projecting and displaying a surveillance camera image on the 3D model described with reference to FIG. 5A. Referring to FIG. 5B, the system 100 may match the user observation time point of the selected view window to the same viewpoint (VPT) as the viewpoint of the surveillance camera, and the image 71 captured by the surveillance camera may be captured in the 3D. You can project onto the model. In this case, the portions 74a and 74b of the image 72 provided by the view window other than the image 71 monitored by the surveillance camera may be represented as a 3D model.

According to the method, the system 100 can match the two-dimensional image 71 captured by the surveillance camera so as to fit the three-dimensional model, so that an object photographed by the surveillance camera is incorrect. For example, the display can be displayed without crushing, blurring or ambiguity, etc., and the intuitiveness of which point in the space of the 3D model corresponds to the image captured by the surveillance camera can be increased.

FIG. 6A is a diagram for describing a method for visualizing a surveillance camera image in a 3D model when the number of surveillance cameras is large. If the number of surveillance cameras is large, reducing the size of the image can be viewed as a whole, but there may be a case where you want to selectively display only a specific area, it is a method for solving this.

Referring to FIG. 6A, the user may only bound the area BD1 to be displayed in the entire image 40 of the virtual camera VCAM0, and the surveillance cameras A1 to A3 in the bounding BD1 area. Only images 41 to 43 may be shown to the user. In this case, images captured by the surveillance cameras A4 to A6 outside the bounding BD1 area are not provided to the user.

FIG. 6B is an example in which the method for visualizing a surveillance camera image in a three-dimensional model according to FIG. 6A is implemented by the system 100 of FIG. 2. The system 100 may display only the screens of the surveillance cameras A1 to A3 desired by the user as shown in FIG. 6B. The display shown in FIG. 6B is particularly useful when the number of surveillance cameras is large.

The user may be bounded BD1 to include desired surveillance cameras A1 to A3 in the entire image 40, and the surveillance cameras A1 to A3 included in the bounding BD1 may monitor. View windows 41 to 43 representing the image may be displayed. If there are a plurality of surveillance cameras included in the bounding BD1, the view window of the surveillance camera selected by using a predetermined scroll SCR1 is determined to be an optimal size in consideration of a predetermined size (a user and a display screen). Display size).

Since the surveillance cameras A4 to A6 are not included in the bounding BD1, the system 100 may not display view windows indicating an image monitored by the surveillance cameras A4 to A6.

The user can select one of the view windows 41-46 shown in FIG. 4A or 6B, and if the user selects one, the selected view window can be displayed full-screen. In this case, the screen provided by the selected view window may be projected on a 3D model that is the same as or similar to the scene to be photographed to provide a user with realism.

FIG. 7A is a diagram illustrating a method for visualizing a surveillance camera image in a case where a 3D model is concentrated in a large area. When three-dimensional models are concentrated in a large area, it is difficult to construct all three-dimensional models, and unnecessary three-dimensional models can make monitoring difficult.

Referring to FIG. 7A, an image processed by the system 100 includes view maps 71 to 74 having predetermined sizes representing a map image 70 and a two-dimensional image monitored by the surveillance cameras B1 to B4. ). Such a display makes it possible to easily determine the positions of the surveillance cameras B1 to B4 and to allow the user to manage the system more easily.

Further, according to the method, on the map image 70, the position of the surveillance cameras B1 to B4, the surveillance direction of the surveillance cameras B1 to B4, and the like of the surveillance cameras B1 to B4 are used. The surveillance areas DET1 to DET4 may be represented, and the far plane may be flexibly represented according to the resolution of the camera (eg, 640 × 480 is represented by 30m).

When the user selects the view window 72, it may appear as shown in FIG. 5A.

This may facilitate the construction of a three-dimensional model in part rather than the entire area.

FIG. 7B is a diagram for illustrating a method for visualizing a surveillance camera image for an area where a 3D model is concentrated when the number of surveillance cameras is large. Referring to FIG. 7B, only the screens of the surveillance cameras B1, B3, and B4 desired by the user may be displayed.

The user may bound BD2 to include desired surveillance cameras B1, B3, and B4 on the screen of FIG. 7A, and the surveillance cameras B1, B3, and B4 included in the bounding BD2 may monitor each other. View windows 71, 73, and 74 representing the image may be displayed. If there are a plurality of surveillance cameras included in the bounding BD2, the view window of the surveillance camera selected using a predetermined scroll SCR2 is determined to be an optimal size in consideration of a predetermined size (a user and a display screen). Display size).

Further, according to the method, on the map image 70, by using the position of the selected surveillance cameras B1, B3, B4, the surveillance direction of the selected surveillance cameras B1, B3, B4, and the like, the surveillance camera B1. , B3, B4) may represent the surveillance areas DET1, DET3, DET4, and the far plane may be flexibly represented according to the resolution of the camera (for example, 640 × 480 is represented by 30m).

Since the surveillance camera B2 is not included in the bounding BD2, the system 100 may not display a view window 72 indicating an image monitored by the surveillance camera B2.

FIG. 8 is a diagram illustrating a method for visualizing a surveillance camera image in a three-dimensional model in a plan view. Referring to FIG. 8, the image processed by the system 100 may be a view window 81 having a predetermined size indicating a plan view-type image 80 and a two-dimensional image monitored by the surveillance cameras C1 to C4. To 84).

In addition, the system 100 may display the surveillance areas DET1 ′ and DET4 ′ monitored by the surveillance cameras C1 to C4 on the map image 80.

9A and 9B are diagrams for explaining an algorithm for detecting an accurate position and a surveillance direction of a surveillance camera on a three-dimensional model based on the captured image.

FIG. 9A illustrates an example of a screen 90 captured by a surveillance camera, and FIG. 9B illustrates an arbitrary point E1, a vanishing point VP1, and a vanishing point VP2 when three-dimensionally considering FIG. 9A. It is a drawing which expresses the passing plane two-dimensionally.

9A and 9B, when the screen 90 is considered in three dimensions, the vanishing point VP1 and the vanishing point VP2 may be known through parallel lines PL1 and parallel lines PL2 perpendicular to each other. Using this, internal parameters of the surveillance camera CAM, such as focal length, image center, etc., can be obtained.

In particular, when the camera has no skew and the pixel is square, it can be simply obtained as in Equation 1 below.

는 이미지 센터의 2차원 벡터, 상기 L은 초점거리,

는 소실점 VP1 2차원 벡터,

는 소실점 VP2의 2차원 벡터, 및 K는 투영 행렬의 형태이다.here,

Is the two-dimensional vector of the image center, L is the focal length,

Is the vanishing point VP1 two-dimensional vector,

Is a two-dimensional vector of vanishing point VP2, and K is in the form of a projection matrix.

Using Equation 1, a three-dimensional vector 901 of any one point (eg, E1) on the screen 90 may be obtained. For example, the inverse of K may be matrix-multiplied by a two-dimensional coordinate value (or two-dimensional vector) of E1 (ie, E1 = (e1, e2)). In this case, for the 3 × 3 matrix multiplication, in the case of the 2-dimensional coordinate value matrix of E1, component 1 may be used at the end as shown in Equation 2 below.

Next, a new three-dimensional coordinate system using the three-dimensional vectors 901 and 904 of the vanishing points VP1 and VP2 and the cross-product vector 903 obtained by cross-sectionalizing the three-dimensional vectors 901 and 904. The intersection of the three-dimensional vector 901 of the arbitrary point (E1) and the focal plane (FPL) of the surveillance camera (CAM) (so-called arbitrary point (E1)) Is projected onto the focal plane FPL, referred to as the origin below) OP.

Next, inputting the distance (id2) of any two points (for example, E1 and E2) on the axis passing through the arbitrary point (E1) and vanishing point (VP1), the origin (OP) ) And the focal length id3 between the surveillance camera CAM.

Next, similarly, the other point E2 except for E1 can also obtain the three-dimensional vector 902 of E2 by the method described above.

Next, when the line 905 parallel to the id2 is passed through the origin OP, the point AD where the line 905 and the three-dimensional vector 902 of E2 meet. Therefore, the distance id1 between the origin OP and the point AD can be obtained.

Next, since id1, id2, and id3 are known, id4 can be obtained using a proportional expression, and thus, it can be known how far the origin OP is located from any point E1.

In this way, the surveillance camera CAM is represented again in a new three-dimensional coordinate system centering on the point E1, and the new three-dimensional coordinate system is positioned in the three-dimensional model according to a predetermined reference or simple user operation. The position and / or surveillance direction of the surveillance camera can be calculated on a three-dimensional model.

In this way, the image captured by the surveillance camera CAM can be seen more accurately where the surveillance camera CAM is located.

Using this algorithm, three-dimensional modeling can be performed inversely from captured images of the plurality of surveillance cameras. For example, in FIG. 9A, one surface of the building may be simply modeled / textured on one plane based on the point E1.

When the surveillance camera CAM is positioned in the three-dimensional model, the newly modeled one-sided information can be used so that the user can position the surveillance camera CAM much more intuitively. It is also possible to more accurately implement virtual three-dimensional models (eg, buildings or terrain) in cyberspace.

The present invention can convert the model represented in the world coordinate system into a virtual camera coordinate system to project a realistic feeling (eg, perspective / depth) of the three-dimensional model, and project it onto an image plane. (projection) can be converted and converted to a regular coordinate system, and finally to a window coordinate system. This can be implemented using OpenGL or DirectX.

The present invention can also be embodied as computer-readable codes on a computer-readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored.

Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like.

The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. And functional programs, codes and code segments for implementing the present invention can be easily inferred by programmers in the art to which the present invention belongs.

Although the present invention has been described with reference to one embodiment shown in the drawings, this is merely exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

100: system, 110: processor,
111: detection module, 112: modeling module,
113: interface, 120: database,
200: electronic device

Claims (15)

Generating a virtual object in which an object in which at least one surveillance camera is located is implemented as a virtual three-dimensional model;
Projecting an image of the at least one surveillance camera onto a plane facing the at least one surveillance camera; And
And adjusting the plane to be perpendicular to a line of sight of a virtual camera, which is a camera virtually set for rendering. The method of claim 1, wherein the method for visualizing the image in the three-dimensional model
Displaying on the virtual object a two-dimensional image photographed by each of the surveillance cameras appearing in the adjusted plane through at least one view window of a predetermined size. Way. The method of claim 2, wherein the method for visualizing the image in the three-dimensional model
And when the user selects one of the at least one view window, projecting the image captured by the corresponding surveillance camera of the selected view window onto the virtual object and outputting the image on the virtual object. . The method of claim 2, wherein the method for visualizing the image in the three-dimensional model
And projecting and outputting an image photographed by the virtual camera and a surveillance camera close to a line of sight or distance on the virtual object and outputting the image. The method of claim 2,
The at least one view window is a method for visualizing the image in the three-dimensional model, the color constituting the at least one view window, or when the at least one view window flickers when a predetermined situation occurs. The method of claim 2, wherein displaying through the at least one view window comprises:
Determining color values of the at least one view window in a predetermined order; And
And moving the at least one view window in a predetermined direction by comparing the color value with a reference value to display the image in a three-dimensional model. The method of claim 2, wherein displaying through the at least one view window comprises:
And displaying the view windows by connecting an annotation line with a corresponding surveillance camera position on the virtual object. The method of claim 2, wherein the displaying through the at least one view window
If a particular area on the virtual object is selected, displaying a view window of surveillance cameras within the selected area. The method of claim 1, wherein generating the virtual object
Generating and rendering the virtual object using wireframe rendering and / or polygon rendering, or alpha value adjusted rendering, to visualize the surveillance region of the surveillance image and surveillance camera. A method for visualizing an image on a three-dimensional model. 10. The method of claim 9,
The polygon rendering method for visualizing an image in a three-dimensional model using a shadow-map algorithm. The method of claim 1, wherein generating the virtual object
Generating a three-dimensional coordinate system using three-dimensional vectors in a specific direction directed to vanishing points of the captured image and an outer vector of the three-dimensional vectors based on the captured images of the at least one surveillance camera; And
Positioning the at least one surveillance camera within the virtual object after calculating the position and / or the surveillance direction of the at least one surveillance camera by positioning the three-dimensional coordinate system on the virtual object according to a predetermined criterion. A method for visualizing an image on a three-dimensional model. A recording medium having recorded thereon a computer program for executing the method according to any one of claims 1 to 11. In a system for visualizing an image on a three-dimensional model,
Create a virtual object that implements an object in which at least one surveillance camera is located as a virtual three-dimensional model, and the planar image facing the at least one surveillance camera in front of the image of the at least one surveillance camera. A processor programmed to project the projection to adjust the plane to be perpendicular to a line of sight of the virtual camera, the camera virtually set for rendering; And
And a database for storing data input and / or output to the system for processing by the processor. The processor of claim 13, wherein the processor is
Create a virtual object that implements an object in which at least one surveillance camera is located as a virtual three-dimensional model, and the planar image facing the at least one surveillance camera in front of the image of the at least one surveillance camera. A modeling module for projecting onto and modeling the plane to be perpendicular to a line of sight of a virtual camera, which is a camera virtually set for rendering;
A detection module for detecting and transmitting data input and / or output to the system for modeling the modeling module; And
And a color buffer for rendering the data modeled by the modeling module as color values. The system of claim 13, wherein the system for visualizing the image in the three-dimensional model
And a display device for displaying a result processed by the processor.

Images