KR20160039447A

KR20160039447A - Spatial analysis system using stereo camera.

Info

Publication number: KR20160039447A
Application number: KR1020140132531A
Authority: KR
Inventors: 이종훈; 천준호; 박성근; 김도윤
Original assignee: 공간정보기술 주식회사
Priority date: 2014-10-01
Filing date: 2014-10-01
Publication date: 2016-04-11

Abstract

The present invention relates to a spatial analysis system using a stereo camera. More particularly, the present invention relates to a spatial analysis system using a stereo camera, in which spatial information obtained by processing a moving object in a stereo image into a three-dimensional image is mapped to a real map, And a spatial analysis system using a stereo camera capable of calculating the position, velocity, direction, area, size, and length of a moving object.

Description

{Spatial analysis system using stereo camera}

The present invention relates to a spatial analysis system using a stereo camera, and more particularly, to a spatial analysis system using a stereo camera, in which spatial information obtained by processing a moving object in a stereo image into a three-dimensional image is mapped to an actual map, The position, speed, direction, area, size, length, and the like of the moving object can be obtained by extracting the object by using the pixel information and the three-dimensional information with respect to the portion corresponding to the moving object, The present invention relates to a spatial analysis system using a stereo camera.

The technique of reconstructing the 3D model from the image is a very old research topic in the field of computer vision.

As a typical technique, there is a technique of restoring three-dimensional information from the relationship of light and shadow (shape from shading), a technique of restoring three-dimensional information from an object surface pattern (shape from texture) (visual hull).

In particular, the visual hull reconstruction method from the contour information is very simple and efficient, and the 3D model can be reconstructed.

However, this technique has a problem in that it can not be restored to the concave area, and a very coarse model is generated when the number of cameras is small.

On the other hand, image processing techniques using stereo images have been actively studied. However, it is a technique for estimating the depth using the binocular parallax generated in images obtained from two cameras, but it is urgent to develop application technology to utilize them.

In addition, as a feature of the existing intelligent CCTV, artificial intelligence recognition has been developed in the fields of human face, object recognition, and object tracking. However, there is a problem in that objects are discriminated only by resolution and pixel brightness.

In the case of conventional fixed CCTV, tracking of objects can not be continued after leaving the surveillance area, and the surveillance area is likely to be changed due to external influences such as age, wind, heat, rain and vibration.

The disadvantage of conventional CCTV is that it is difficult to correct the distortion, and the precise position, length, height, size, and orientation of the camera and object are unknown.

Korean Registered Patent No. 10-1290197 (Feb.

The spatial analysis system using the stereo camera of the present invention is intended to solve the problems occurring in the related art as described above. The spatial information obtained by processing a moving object in a stereo image into a three-dimensional image is mapped to a real map, It is possible to perform object tracking by extracting an arbitrary range, extracting an object using pixel information and three-dimensional information for a portion corresponding to a range, implementing a spatial function for the object, Direction, area, size, length, and so on.

Another object of the present invention is to enable various statistics using spatial information.

In order to solve the above problems, a spatial analysis system using a stereo camera according to the present invention includes a stereo image acquisition unit (100) for acquiring a stereo image;

A line corrector 200 for correcting lines of the left image and the right image;

An epipolar section 300 for correcting the vertical parallax of the three-dimensional stereo image based on the left image;

A depth map generator 400 for acquiring left and right stereo images and extracting matching pointer attributes to automatically generate a 3D depth map;

A map map database 500 storing map information;

A moving object analyzer 600 for extracting spatial information from the map map database and then analyzing the position, velocity, and direction of the moving object by mapping an actual moving object to an actual space map using a spatial function;

A setting information obtaining unit 700 for obtaining request variable information set by a user among the length, distance, area, and volume of the moving object;

A request variable analyzing unit 800 for analyzing a variable value corresponding to request information among the length, distance, area, and volume of the moving object based on the request information acquired by the setting information obtaining unit;

And a result display unit 900 for displaying a result value by the moving object analysis unit and a variable value analyzed by the request variable analysis unit as a stereo image and a depth map, thereby solving the problems of the present invention .

According to the present invention, spatial information obtained by processing a moving object in a stereo image into a three-dimensional image is mapped to a real map, and then an arbitrary range is classified in space, and pixel information and three-dimensional information It is possible to calculate the position, speed, direction, area, size, length, etc. of the moving object by extracting the object and implementing the spatial function for the object so that the object tracking can be performed. It provides the effect of analyzing the position and the movement path without any.

In addition, various statistics can be made by using spatial information, and it is possible to use the data as a data when introducing a child protection zone and a traffic light since the moving route of people and vehicles can be known.

1 is an overall block diagram of an intelligent 3D cctv camera image processing apparatus according to an embodiment of the present invention.
FIG. 3 is a diagram illustrating an example of executing a depth map, FIG. 4 is an example of an administrator setting a moving object on an actual screen, FIG. 5 is an example of object tracking .

Hereinafter, a spatial analysis system using a stereo camera according to the present invention will be described in detail with reference to the accompanying drawings.

1 is an overall block diagram of an intelligent 3D cctv camera image processing apparatus according to an embodiment of the present invention.

As shown in FIG. 1, a spatial analysis system 1000 using a stereo camera includes a stereo image acquisition unit 100; A line correction unit 200; An epi polar portion 300; A depth map generating unit 400; A map map database 500; A moving object analysis unit 600; A setting information obtaining unit 700; A request variable analysis unit 800; And a result display unit 900.

The stereo image acquisition unit 100 acquires a stereo image from the stereo camera 10.

The stereo image means a left image and a right image captured by two cameras.

The line correction unit 200 corrects the lines of the left image and the right image as shown in FIG.

In other words, if shaking occurs, an error may occur during data processing, so correction is performed.

Thereafter, the epipolar unit 300 corrects the vertical parallax of the 3D stereo image based on the left image.

As shown in FIG. 3, the depth map generator 400 acquires a left and right stereo image and extracts a matching pointer attribute to automatically generate a 3D depth map.

In this case, the position of the moving object is mapped to the actual space map using the map map database in the present invention.

For this purpose, a map map database 500 storing map information is constructed. The moving object analyzing unit 600 extracts spatial information from the map map database, and then, using the spatial function, Mapping, and analyzing the position, speed, and direction of the moving object.

It is easy to analyze because it has video information moving in real time rather than image in speed and direction.

Since the actual stereo image is obtained from the camera having the fixed position value, the moving object can be identified and mapping to the map becomes possible.

As described above, since the movement route can be known, it becomes possible to designate a child protection zone, for example, or to utilize it as evidence data before introducing a traffic light.

That is, it can be utilized as various statistical information using spatial information.

In addition, since spatial information is matched with a map, object tracking is possible by implementing a spatial function, so that it can be utilized for tracking a moving object.

As shown in FIG. 4, the setting information obtaining unit 700 can specify a moving object by the user and check the length, distance, area, volume, and the like of the moving object.

For this purpose, the request variable information set by the user among the length, distance, area, and volume of the moving object is obtained.

The request variable analyzing unit 800 analyzes a variable value corresponding to request information among the length, distance, area, and volume of the moving object based on the request information acquired by the setting information obtaining unit.

That is, when a point and a point are selected on the screen and they are connected, the length of the moving object can be obtained, the distance can be obtained, and the area can be calculated.

The result display unit 900 displays a result value of the moving object analyzing unit and a variable value analyzed by the request variable analyzing unit as a stereo image and a depth map.

That is, as shown in Fig. 5, the image is displayed.

It is possible to perform object tracking by mapping a spatial information obtained by processing a moving object in a stereo image into a three-dimensional image through an arrangement and an operation of the present invention with an actual map and implementing a spatial function, , Area, size, length, and so on, it is possible to analyze the position and movement path regardless of the background object even if there is an obstacle.

The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

100: stereo image acquisition unit
200: line correction unit
300: epipolar part
400: depth map generating unit
500: map map div
600: Moving Object Analysis Unit
700: Setting information acquiring unit
800: request variable analysis unit
900: Result display section

Claims

In a spatial analysis system using a stereo camera,
A stereo image acquiring unit (100) for acquiring a stereo image;
A line corrector 200 for correcting lines of the left image and the right image;
An epipolar section 300 for correcting the vertical parallax of the three-dimensional stereo image based on the left image;
A depth map generator 400 for acquiring left and right stereo images and extracting matching pointer attributes to automatically generate a 3D depth map;
A map map database 500 storing map information;
A moving object analyzer 600 for extracting spatial information from the map map database and then analyzing the position, velocity, and direction of the moving object by mapping an actual moving object to an actual space map using a spatial function;
A setting information obtaining unit 700 for obtaining request variable information set by a user among the length, distance, area, and volume of the moving object;
A request variable analyzing unit 800 for analyzing a variable value corresponding to request information among the length, distance, area, and volume of the moving object based on the request information acquired by the setting information obtaining unit;
And a result display unit (900) for displaying a result value by the moving object analysis unit and a variable value analyzed by the request variable analysis unit as a stereo image and a depth map. Spatial Analysis System.

The method according to claim 1,
The moving object analyzing unit,
Wherein the position and movement path of the moving object can be analyzed regardless of the background object even if there is an obstacle.

The method according to claim 1,
The moving object analyzing unit 600,
Wherein spatial information is mapped to an actual map and then object tracking is performed by implementing a spatial function.