KR20170100204A - Apparatus and method for target tracking using 3d bim - Google Patents

Abstract

Disclosed are a device and a method for tracking a target by using three-dimensional spatial information, which digitize information of a three-dimensional space and a target by mapping image information captured by capturing devices and three-dimensional spatial information (BIM). According to the present invention, the device for tracking a target by using three-dimensional spatial information comprises: a transformation matrix calculating part mapping a plurality of capturing devices and the three-dimensional spatial information corresponding to a real space captured by the capturing devices, and calculating a transformation matrix; a target detecting part detecting the target included in the image information captured by at least one capturing device among the plurality of capturing devices, and extracting characteristic information of the detected target; a three-dimensional coordinate extracting part extracting a three-dimensional virtual spatial coordinate of the detected target by using the transformation matrix; and a path tracking part comparing a previous frame of the image information with a current frame, and tracking a moving path of the target.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and method for tracking a target using 3D spatial information,

The present invention relates to a target tracking technique using three-dimensional spatial information and image information. More particularly, the present invention relates to a technique for detecting a target included in image information acquired from a photographing apparatus installed in a building, ≪ / RTI >

Today, CCTV is widely applied to crime prevention, disaster prevention, facility management, and traffic system management. CCTV control center for crime prevention and disaster prevention receives images from hundreds of cameras. Therefore, it is difficult for a small number of control personnel to control all cameras. Image-based intelligent CCTV technology is being developed to solve these problems and to recognize crime and disaster situations by using image processing and image recognition technology to efficiently control CCTV.

In order to implement an image-based intelligent CCTV control system, object detection, object tracking, and situation recognition technology are required. Among them, object tracking, which tracks the targets included in CCTV images, is a mature technology that commercial products have been released.

Recently, a technique of mapping CCTV images to BIM information and projecting them in a three-dimensional virtual space has also been developed. For example, commercial tools like Autodesk's Revit are typical.

However, when a CCTV image is projected onto a virtual three-dimensional space using BIM instead of on-screen coordinates, a technique of storing the moving path of the target in three-dimensional coordinates has been studied recently.

Until now, the police who use CCTV video and investigate the epidemiological investigation to identify the infection route have been able to visually confirm multiple CCTV images one by one and create a path through the hand. Therefore, it takes a lot of time to check the CCTV image, and there is a problem that the confirmation result of the CCTV image is not digitized.

Therefore, there is an urgent need for a technique for automatically tracking a target from a CCTV image and digitizing the movement path of the target in three dimensions.

Korean Patent Laid-Open No. 10-2015-0132968, published on November 27, 2015 (name: Multiple goal tracking system and method)

An object of the present invention is to automatically track a target included in image information photographed by a photographing apparatus and digitize the moving path of the target in three dimensions by expressing the moving path of the target as x, y coordinates and the number of layers of floor information.

It is also an object of the present invention to predict the anticipated movement path of the target in the obscured section and to digitize it in three dimensions when the tracking of the target is missed.

In addition, an object of the present invention is to digitize information of a three-dimensional space and a target by mapping image information and three-dimensional spatial information (BIM) photographed by photographing apparatuses.

It is also an object of the present invention to enable digitalized stored data to be reconstructed at any time in multiple angles for access and utilization.

It is also an object of the present invention to dramatically reduce the amount of data required in tracking a target by digitizing and storing the 3D virtual space coordinate and the movement path of the target.

According to an aspect of the present invention, there is provided a target tracking apparatus using three-dimensional spatial information, comprising: a plurality of photographic apparatuses; mapping three-dimensional spatial information corresponding to an actual space photographed by the photographic apparatuses, A target detecting unit detecting a target included in the image information captured by at least one of the plurality of photographing apparatuses and extracting feature information of the detected target from the plurality of photographing apparatuses, A 3D coordinate extractor for extracting a 3D virtual space coordinate of the detected object, and a path traceer for comparing a current frame with a previous frame of the image information to track a movement path of the target.

At this time, the conversion matrix operation unit can map the image information captured by the photographing apparatus and the three-dimensional spatial information using markers attached to the actual space.

At this time, the transformation matrix operation unit may restore the image information into three-dimensional solid information using a coupled-line camera technique, and map the three-dimensional solid information to the three-dimensional spatial information .

At this time, the 3D coordinate extraction unit may extract the 3D virtual space including at least one of the ID information of the target, the 3D virtual space coordinate information, the 3D virtual space layer number information, The phase coordinate can be extracted.

In this case, when the tracking of the target is missed, the target detecting unit may detect the expected behavior radius of the target, and the target detecting unit may redetect the target using the feature information of the target.

At this time, the predicting unit can search for an expected behavior radius of the target using a prediction algorithm including at least one of a social path planning and an extended Kalman filter.

At this time, the prediction unit may predict a predicted movement path of the target using a prediction algorithm including at least one of the social path planning and the extended Kalman filter.

At this time, the target detection unit compares the feature information of the first target included in the first image information with the feature information of the second target included in the second image information, and outputs the feature information of the first target and the second target It is possible to determine whether or not the target is the same target.

At this time, if the first target and the second target are judged to be the same target, the target detecting unit may handover the first target to the second video apparatus that has captured the second video information.

At this time, the path tracing unit may display the traced movement path of the target on a two-dimensional map.

According to another aspect of the present invention, there is provided a target tracking method performed by a target tracking apparatus using three-dimensional spatial information, including: a step of detecting three-dimensional spatial information corresponding to an actual space photographed by a plurality of photographing apparatuses and photographing apparatuses Calculating a transformation matrix by using the transformation matrix, detecting a target included in the image information captured by at least one of the plurality of photographing apparatuses, extracting feature information of the detected target, Extracting coordinates of the 3D virtual space of the target detected using the transformation matrix, and comparing the current frame with the previous frame of the image information to track the movement path of the target.

According to the present invention, the target included in the image information photographed by the photographing apparatus can be automatically tracked, and the moving path of the target can be digitized in three dimensions by expressing the x, y coordinates and the layer number information of the bottom.

In addition, according to the present invention, when the tracking of the target is missed, the predicted moving path of the target in the obscured section can be predicted and digitized into three dimensions.

In addition, according to the present invention, it is possible to digitize the three-dimensional space and the target information by mapping the image information photographed by the photographing apparatuses and the three-dimensional spatial information (BIM).

In addition, according to the present invention, digitized and stored data can be reconfigured at various angles at any time for access and utilization.

Also, according to the present invention, it is possible to dramatically reduce the amount of data required for tracking a target by digitizing and storing the 3D virtual space coordinates and the movement path of the target.

1 is a block diagram illustrating a configuration of a target tracking apparatus using three-dimensional spatial information according to an embodiment of the present invention.
2 is a flowchart illustrating a method of tracking a target using three-dimensional spatial information according to an embodiment of the present invention.
3 is a view showing three-dimensional spatial information of an actual space according to an embodiment of the present invention.
4 is a view showing a marker attached to a floor of an actual building according to an embodiment of the present invention.
5 is a view for explaining a photographing apparatus for locating a marker according to an embodiment of the present invention.
6 is a view for explaining the positions of the marker and the photographing apparatus in the three-dimensional spatial information according to the embodiment of the present invention.
7 is a view showing bottom coordinate pixels of a target according to an embodiment of the present invention.
FIG. 8 is a view showing 3D virtual space coordinates of a target in a virtual space according to an embodiment of the present invention.
FIG. 9 is a diagram showing a two-dimensional map showing a moving path of a target according to an embodiment of the present invention.
FIG. 10 is a diagram for explaining a process of searching for an expected behavior radius in a tracking missing interval according to an embodiment of the present invention.
11 is a view showing a target captured in image information according to an embodiment of the present invention.
12 is a diagram showing a target displayed on a three-dimensional space map according to an embodiment of the present invention.
13 is a diagram for explaining a process of storing a moving route of a target according to an embodiment of the present invention.

The present invention will now be described in detail with reference to the accompanying drawings. Hereinafter, a repeated description, a known function that may obscure the gist of the present invention, and a detailed description of the configuration will be omitted. Embodiments of the present invention are provided to more fully describe the present invention to those skilled in the art. Accordingly, the shapes and sizes of the elements in the drawings and the like can be exaggerated for clarity.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating a configuration of a target tracking apparatus using three-dimensional spatial information according to an embodiment of the present invention.

1, the target tracking apparatus 100 using three-dimensional spatial information includes a conversion matrix calculation unit 110, a target detection unit 120, a 3D coordinate extraction unit 130, a path tracking unit 140, (150).

First, the transformation matrix operation unit 110 maps the three-dimensional spatial information corresponding to the angle of view of a plurality of photographing apparatuses and the actual space photographed by the photographing apparatuses, and calculates a transformation matrix. At this time, the transformation matrix operation unit 110 can reconstruct the image information into three-dimensional solid information using the coupled-line camera technique, and map the three-dimensional solid information to the three-dimensional spatial information.

Also, the conversion matrix operation unit 110 can calculate a matrix for mapping the angle of view of the image information photographed by the photographing apparatus and the three-dimensional spatial information using markers attached to the actual space.

Then, the target detecting unit 120 detects the target included in the image information and extracts the feature information of the detected target. At this time, the target detecting unit 120 compares the feature information of the first target included in the first image information with the feature information of the second target included in the second image information, and determines whether the first target and the second target are the same target Can be determined.

If it is determined that the first target and the second target are the same target, the target detecting unit 120 may hand over the first target to the second video apparatus that has captured the second video information.

Further, if the tracking of the target is missed and the predicting unit 150 searches for the expected behavior radius of the target, the target detecting unit 120 can re-detect the target using the feature information of the target.

Next, the 3D coordinate extraction unit 130 extracts 3D virtual space coordinates of the detected target using the transformation matrix.

In addition, the 3D coordinate extraction unit 130 extracts coordinates of the 3D virtual space including at least one of the ID information of the target, the 3D virtual space coordinate information, the 3D virtual space layer number information, can do.

The path trace unit 140 compares the previous frame of the image information with the current frame and tracks the movement path of the target.

The path tracing unit 140 can display the trajectory of the traced target on a two-dimensional map which is the bottom surface of the three-dimensional space map. At this time, the route tracking unit 140 can track the movement path of the target by comparing the t-1 frame, which is the previous frame, with the t frame, which is the current frame.

Finally, when the prediction unit 150 misses the tracking of the target, it searches the expected action radius of the target.

Then, the prediction unit 150 can use the prediction algorithm including at least one of the social path planning and the extended Kalman filter to search for the expected behavior radius of the target. In addition, the prediction unit 150 can predict a predicted movement path of the target using a prediction algorithm including at least one of a social path planning and an extended Kalman filter.

For convenience of explanation, the prediction unit 150 has been described as using the prediction algorithm including at least one of the social path planning and the extended Kalman filter to predict the expected movement radius of the target or the expected movement route of the target. However, the present invention is not limited to this, and the prediction unit 150 may estimate the target using a prediction algorithm that is commonly used, such as a Partial filter, a Bayesian filter, and a Markov chain Monte Carlo. The behavior radius and the predicted movement path can be predicted.

The target tracking apparatus 100 using three-dimensional spatial information includes a storage unit (not shown) for storing at least one of coordinates of a plurality of photographing devices, three-dimensional spatial information corresponding to a real space, Time).

At this time, the storage unit digitally stores coordinates and movement path of the 3D virtual space of the target, thereby dramatically reducing the amount of data required for tracking the target.

In addition, when the target tracking apparatus 100 using three-dimensional spatial information does not include a storage unit, information on a plurality of photographing apparatuses, three-dimensional spatial information corresponding to an actual space, and 3D virtual And spatial coordinates.

Hereinafter, a target tracking method performed by the target tracking apparatus using three-dimensional spatial information according to an embodiment of the present invention will be described in detail with reference to FIG. 2 through FIG.

2 is a flowchart illustrating a method of tracking a target using three-dimensional spatial information according to an embodiment of the present invention.

First, the target tracking apparatus 100 maps the three-dimensional spatial information corresponding to the photographing apparatus and the actual space, and calculates a transformation matrix (S210).

The target tracking apparatus 100 performs a preliminary operation for mapping three-dimensional spatial information of an actual space taken by a photographing apparatus such as a CCTV and the photographing apparatus. Here, the pre-task refers to a task performed to determine the position of the target captured in the image information on the three-dimensional virtual space using the BIM.

The target tracking apparatus 100 performs a preliminary operation to calculate a transformation matrix on the 3D image and the image information. At this time, the image information and the three-dimensional spatial information can be mapped using a conventional mapping technique.

3 is a view showing three-dimensional spatial information of an actual space according to an embodiment of the present invention.

The three-dimensional space information as shown in FIG. 3 includes information about the walls and the floor of the building. The target tracking apparatus 100 acquires three-dimensional spatial information through the BIM of the actual space in which the photographing apparatus is installed, Dimensional space information can be acquired. The target tracking device 100 can digitize and store three-dimensional spatial information.

4 is a view showing a marker attached to a floor of an actual building according to an embodiment of the present invention.

As shown in FIG. 4, in order to improve the accuracy of the floor position, a marker 400 is attached to a floor of an actual building to calibrate the imaging apparatus and a transfer matrix ).

5 is a view for explaining a photographing apparatus for locating a marker according to an embodiment of the present invention.

5, the photographing apparatus 500 photographs a space including the markers 400 attached to the actual floor. The target tracking apparatus 100 confirms where the marker 400 is located on the image information captured by the photographing apparatus 500. [

The target tracking apparatus 100 may be configured to detect a square on the screen of the photographing apparatus 500 by using a CLC (Coupled Line Camera) technique in order to map three-dimensional spatial information and image information based on the marker 400. [ Into a three-dimensional digital space, and computes a transformation matrix.

Equation (1) shows a transformation matrix.

Here, T denotes a transformation matrix, P denotes pixel coordinates of image information, and M denotes coordinates in a three-dimensional space.

That is, the transformation matrix T means a matrix for mapping the pixel coordinates (Px, Py) of the image information to the three-dimensional space (id, x, t, f, t). In this case, id denotes an identifier of the target, x denotes an x coordinate in a three-dimensional space, y denotes a y coordinate in a three-dimensional space, f denotes a number of building layers in a three- And means a time when the photographing apparatus photographed the corresponding image.

6 is a view for explaining the positions of the marker and the photographing apparatus in the three-dimensional spatial information according to the embodiment of the present invention.

6, the target tracking apparatus 100 restores the photographing apparatus 500 and the marker 400 on a three-dimensional space using a CLC (Coupled Line Camera) technique.

For convenience of explanation, the target tracking apparatus 100 has been described as mapping three-dimensional spatial information with one photographing apparatus. However, the present invention is not limited to this, and the target tracking apparatus 100 may be connected to a plurality of photographing apparatuses, The device and the three-dimensional spatial information can be mapped.

Next, the target tracking apparatus 100 detects a target included in the image information (S220).

Here, the image information means an image captured by the photographing device mapped with the three-dimensional spatial information in step S210. The target tracking apparatus 100 secures the image information of the photographing apparatus to be subjected to epidemiological investigation at the time of actual operation after completion of the preliminary work. Then, the target included in the image information is extracted and tracked to obtain the coordinates of the 3D virtual space. At this time, the target tracking device 100 can automate and process the operation of calculating and storing the coordinates on the virtual space.

7 is a view showing bottom coordinate pixels of a target according to an embodiment of the present invention.

7, the target tracking apparatus 100 tags the target 700 included in the image information and obtains the bottom coordinate pixels Px and Py of the target 700. As shown in FIG.

At this time, the target tracking device 100 receives the tagging of the target 700 manually from the user, or the target tracking device 100 automatically learns the database of the person or object to detect a person or object included in the image information The target 700 can be detected.

In addition, the target tracking apparatus 100 may compare (t-1) a previous frame with a current t frame to detect (tag) the occurrence of a difference between the two frames as the target 700.

Then, the target tracking device 100 extracts feature information of the tagged target (S230).

When the target is a person, the target tracking device 100 may extract the dressing information including the key of the target, the color of the clothes, the kind of clothes, and the like as the feature information and store the information.

In addition, the target tracking apparatus 100 extracts 3D virtual space coordinates of the detected target using the transformation matrix (S240).

FIG. 8 is a view showing 3D virtual space coordinates of a target in a virtual space according to an embodiment of the present invention.

As shown in FIG. 8, the target tracking apparatus 100 calculates and stores 3D virtual space phase coordinates (ID1, Vx, Vy, Vf, t) of the tagged target 800. At this time, the 3D virtual space coordinate of the target 800 includes an identifier ID1 of the target, an x coordinate Vx on the three-dimensional space, a y coordinate Vy on the three-dimensional space, a number of building layers Vf on the three- And the time (t) at which the photographing apparatus photographed the image.

Finally, the target tracking device 100 tracks the movement path of the target and displays the traced movement path on the two-dimensional map (S250).

FIG. 9 is a diagram showing a two-dimensional map showing a moving path of a target according to an embodiment of the present invention.

As shown in Fig. 9, the target tracking device 100 displays the track 900 of the tracked target on a three-dimensional map which is the bottom surface of the three-dimensional space map.

At this time, the target tracking apparatus 100 compares the t-1 frame and the t frame around the bottom coordinate pixel of the target at the time of tagging, tracks the movement path 900 of the target, and displays it on the two-dimensional map .

Hereinafter, a method for tracking a target in a tracking missing interval according to an embodiment of the present invention using a three-dimensional spatial information according to an embodiment of the present invention will be described in detail with reference to FIGS. 10 to 13. FIG.

FIG. 10 is a diagram for explaining a process of searching for an expected behavior radius in a tracking missing interval according to an embodiment of the present invention.

10, the target tracking apparatus 100 may perform a social path planning (SPP) when an occlusion occurs in a target 800 being tracked and a tracking missing interval 1000 occurs, Technique and an Extended Kalman Filter (EKF) to search for an expected behavior radius of the target 800. The predictive action radius of the target 800 is determined by using a predictive algorithm that includes at least one of the following:

Here, the social path planning (SPP) technology means a human action simulation path planning technology. In other words, the social path planning (SPP) is a technique for planning a route to a place where a person can move on a two-dimensional map.

The social path planning (SPP) technique assumes human characteristics that humans generally want to travel to the shortest distance and prefer a less directional path with a certain distance from the obstacle. And we use these human characteristics to plan the route.

And the extended Kalman filter (EKF) predicts the area that can move with time considering the human moving speed under the planned path.

For convenience of explanation, the target tracking apparatus 100 predicts at least one of an expected movement radius of a target and an expected movement route of the target using a prediction algorithm including at least one of a social path planning and an extended Kalman filter The present invention is not limited thereto and the target tracking apparatus 100 may be configured to detect a target object using a prediction algorithm that is commonly used, such as a Partial filter, a Bayesian filter, and a Markov chain Monte Carlo At least one of the expected behavior radius and the predicted movement path of the vehicle can be predicted.

Next, the target tracking apparatus 100 re-detects the target within the expected action radius using the feature information including the key of the target, the color of the clothes, and the like. Then, the re-detected target is mapped with the target before the occurrence of occlusion.

If the tracking of the track being tracked is missed and the new target is detected in the image information, the target tracking device 100 can tag and track the new target.

11 is a view showing a target captured in image information according to an embodiment of the present invention.

As shown in FIG. 11, the target tracking apparatus 100 can detect a moving person or object in the image information as the target 700, and tag the detected target 700.

At this time, the target tracking device 100 may display a new target differently from an existing target so that the manager can later confirm the tagging of the new target.

12 is a diagram showing a target displayed on a three-dimensional space map according to an embodiment of the present invention.

12, the target tracking apparatus 100 can calculate 3D virtual space coordinates (ID1, Vx, Vy, Vf, t) of the tagged target 800 and display it on a three-dimensional space map have.

13 is a diagram for explaining a process of storing a moving route of a target according to an embodiment of the present invention.

The target tracking device 100 may track the target 800 by a preset time unit or a time unit input by the user or may track the target 800 in real time to generate a movement path. Then, the target tracking device 100 can display the generated movement path on the three-dimensional space map.

In addition, when the target tracking apparatus 100 receives image information from a plurality of photographing apparatuses, the target tracking apparatus 100 compares the feature information of the targets included in the plurality of image information and recognizes the target information as one target.

The target tracking apparatus 100 can recognize a target included in the images input from the plurality of photographing apparatuses as one target by using the movement path of the target tagged on the three-dimensional space.

That is, the target tracking apparatus 100 can determine whether the targets included in the image information corresponding to the plurality of photographing apparatuses are the same target, using the travel path of the target and the feature information of the target.

Further, the target tracking apparatus 100 may handover the target tagged in the first image photographed from the first photographing apparatus to the second photographing apparatus. When the target tracking device 100 handover a tagged target and the switching between screens is inaccurate, the target tracking device 100 can predict an expected path of the target using the extended Kalman filter.

For convenience of explanation, the target tracking apparatus 100 has been described as tracking a single target, but the present invention is not limited thereto. The target tracking apparatus 100 may tag and track each of a plurality of targets, . ≪ / RTI >

As described above, the target tracking apparatus and method using three-dimensional spatial information according to the present invention can automate target tracking using a photographing device such as CCTV, convert information about target tracking into digital data, can do. It is also possible to automate the epidemiological investigation using the image information, thereby shortening the time for confirming the image information of the investigator.

In addition, in the past, it took a long time for a person to check a plurality of pieces of video information one by one. On the other hand, the apparatus and method for tracking a target using three-dimensional spatial information according to the present invention can simultaneously process a plurality of image information, thereby shortening the irradiation time.

Since the image information input from the plurality of photographing apparatuses can be integrated and provided as one information to the user, the readability is improved, the user rotates the viewing angle based on the three-dimensional map, So that the user can view the image information.

As described above, the configuration and method of the embodiments described above can be applied to a target tracking apparatus and method using three-dimensional spatial information according to the present invention. All or some of the embodiments may be selectively combined.

100: Target tracking system using 3D spatial information
110: conversion matrix operation unit
120: target detection unit
130: 3D coordinate extracting unit
140: Path trace section
150:
400: Marker
500: photographing apparatus
700: target included in image information
800: Target in 3D virtual space
900: Travel route
1000: Missing trace section

Claims (20)

A conversion matrix operation unit for mapping a plurality of photographing apparatuses and three-dimensional spatial information corresponding to an actual space photographed by the photographing apparatuses to calculate a conversion matrix,
A target detecting unit that detects a target included in the image information captured by at least one of the plurality of photographing apparatuses and extracts the detected feature information of the target,
A 3D coordinate extracting unit for extracting coordinates of the 3D virtual space of the target detected using the transformation matrix,
And a path tracing unit for comparing a previous frame of the image information with a current frame to track the moving path of the target. The method according to claim 1,
Wherein the conversion matrix calculator comprises:
And mapping the image information captured by the photographing apparatus and the three-dimensional spatial information using a marker attached to the actual space. 3. The method of claim 2,
Wherein the conversion matrix calculator comprises:
Dimensional space information by using a three-dimensional space information, and restoring the image information into three-dimensional space information by using a line camera pair technique (Coupled Line Camera) Tracking device. The method according to claim 1,
The 3D coordinate extracting unit may extract,
The 3D virtual space coordinate including at least one of ID information of the target, 3D virtual space coordinate information, 3D virtual space layer number information, and time information of the image information is extracted. Target Tracking System Using 3D Spatial Information. The method according to claim 1,
Further comprising a predictor for searching for an expected behavior radius of the target when the tracking of the target is missed,
Wherein the target detection unit comprises:
Dimensional space information for re-detecting the target using the feature information of the target. 6. The method of claim 5,
The predicting unit,
Wherein the predicted action radius of the target is searched using a prediction algorithm including at least one of a social path planning and an extended Kalman filter. The method according to claim 6,
The predicting unit,
Wherein the predicted movement path of the target is predicted using a prediction algorithm including at least one of the social path planning and the extended Kalman filter. The method according to claim 1,
Wherein the target detection unit comprises:
The feature information of the first target included in the first image information is compared with the feature information of the second target included in the second image information to determine whether the first target and the second target are the same target Target Tracking System Using 3D Spatial Information. 9. The method of claim 8,
Wherein the target detection unit comprises:
And when the first target and the second target are determined to be the same target, the first target is handed over to the second video device that has captured the second video information. The method according to claim 1,
The path-
And displaying the traced movement path of the target on a two-dimensional map. A target tracking method performed by a target tracking apparatus using three-dimensional spatial information,
Mapping the three-dimensional spatial information corresponding to the actual space photographed by the plurality of photographing apparatuses and the photographing apparatuses to calculate a transformation matrix,
Detecting a target included in image information photographed by at least one photographing apparatus among the plurality of photographing apparatuses,
Extracting feature information of the detected target;
Extracting 3D virtual space coordinates of the target detected using the transformation matrix, and
And comparing the current frame with a previous frame of the image information to track the movement path of the target. 12. The method of claim 11,
Wherein the step of computing the transformation matrix further comprises:
And mapping the image information photographed by the photographing device and the three-dimensional spatial information using a marker attached to the actual space. 13. The method of claim 12,
Wherein the step of computing the transformation matrix further comprises:
Dimensional space information by using a three-dimensional space information, and restoring the image information into three-dimensional space information by using a line camera pair technique (Coupled Line Camera) Tracking method. 12. The method of claim 11,
Wherein the step of extracting the 3D virtual space coordinate comprises:
The 3D virtual space coordinate including at least one of ID information of the target, 3D virtual space coordinate information, 3D virtual space layer number information, and time information of the image information is extracted. Target Tracking Method Using 3D Spatial Information. 12. The method of claim 11,
If the tracking of the target is missed, searching for an expected behavior radius of the target, and
Further comprising the step of re-detecting the target using the feature information of the target. 16. The method of claim 15,
Wherein the step of searching for an expected behavior radius of the target comprises:
Wherein the predicted action radius of the target is searched using a prediction algorithm including at least one of a social path planning and an extended Kalman filter. 17. The method of claim 16,
Wherein the step of searching for an expected behavior radius of the target comprises:
Wherein the predicted movement path of the target is predicted using a prediction algorithm including at least one of the social path planning and the extended Kalman filter. 12. The method of claim 11,
The feature information of the first target included in the first image information is compared with the feature information of the second target included in the second image information to determine whether the first target and the second target are the same target Wherein the method further comprises the steps of: 19. The method of claim 18,
Further comprising the step of, when it is determined that the first target and the second target are the same target, handing over the first target to a second video apparatus that has captured the second video information, Target tracking method. 12. The method of claim 11,
And displaying the traced movement path of the target on a two-dimensional map.

Images