KR20140135368A - Crossroad imaging system using array camera - Google Patents

Abstract

The present invention relates to a crossroad imaging system using an array type camera. A crossroad and a connection road connected to the crossroad are photographed, respectively. A high magnification lens is installed in a remote connection road. A low magnification lens is installed in a short connection road. Images obtained by the cameras are synthetized by a panorama method. Provided is an array type camera which can obtain the image of the same quality in each connection road.

Description

Technical Field [0001] The present invention relates to an intersection imaging system using an array camera,

The present invention relates to an intersection photographing system using an array camera. More specifically, an array camera for photographing each road connected to an intersection S and an intersection S is installed so that an intersection S, And an array camera for acquiring an undistorted intersection image by combining the images obtained by correcting the distortion caused by the distance between the installation position of the array camera and the subject, And an intersection photographing system.

The intersection is a point where at least two roads cross each other, and the quantity increases in proportion to the increase of the road. The increase of the intersection serves as a cause of the vehicle stagnation.

In addition, intersections have a characteristic in which vehicles having different motion vectors or conflicts between vehicles and pedestrians are crowded to cause various traffic accidents, as well as frequent traffic violation vehicles such as signals and tail waggles.

Accordingly, there have been various studies on a traffic system for detecting an accident cause and an abuser or providing a traffic flow (congestion situation) at an intersection when a traffic accident occurs by using an image taken at an intersection.

Conventionally, however, a camera installed at one side of an intersection is configured to photograph an intersection, and such a configuration can acquire an image for an intersection S, but has a limitation that an image for roads connected to an intersection can not be obtained . That is, the conventional photographing system can detect the traffic situation of the intersection through the photographed image, but can not detect the congestion state of the connected roads connected to the intersection.

In addition, conventionally, since a camera installed at a predetermined height on the ground is configured to photograph an intersection, a distortion in a photographed image due to a difference in height between the camera and a subject, that is, a photographing angle of the camera, occurs severely.

1 is an actual photograph showing a conventional intersection image.

The intersection image 100 of FIG. 1 is an image obtained by a camera installed at a position higher than the road surface on one side of an intersection.

In addition, as shown in FIG. 1, the intersection image 100 varies in size depending on the distance from the camera to the subject in the image.

That is, the vehicle lengths d, d ', and d' 'are small in proportion to the distance from the camera in the images of the vehicles 101, 102, and 103 having similar vehicle lengths in the actual space .

Also, it can be seen that the vehicle length of the vehicle 101 photographed at a position adjacent to the camera in the image is the largest.

Also, the conventional intersection image can acquire an image for a predetermined intersection area S, but the image for the roads connected to the intersection S has a limit that can not be obtained. Accordingly, It can not judge the congestion situation of the connected roads.

Thus, conventionally, an intersection image can not provide accurate traffic information because it can not acquire an image of roads connected to an intersection, and a distortion phenomenon occurs severely, so that even if a traffic accident occurs, It has a disadvantage that it is not easy to discriminate.

2 is an actual image of an intersection disclosed in Korean Patent Laid-Open No. 10-2011-0105527 (entitled " Integrated Traffic Light Management Apparatus and Method Including an Intersection Accident Recording and Imaging Apparatus ").

The signal light control system of FIG. 2 is intended to more accurately grasp the situation at an intersection traffic accident while reducing manufacturing costs by integrating and linking various devices such as a signal controller, a video camera, and a video recording device.

2, since the image 200 is installed at a position higher than the ground, that is, the camera photographs the intersection with the photographing angle, As described above, the subject adjacent to the position of the camera in the image 200 is large, and the subject at a distance from the camera is photographed small, which has a limitation in not overcoming the problem of FIG.

In addition, the intersection image 200 is obtained by taking the intersection area S as a focal point and photographing the image of the intersection area relatively accurately, but the roads 201-1, 201-2, 201-3 connected to the intersection S, , And (201-4), it can not provide traffic information such as a traffic congestion situation.

In order to solve such a problem, the camera may be installed at a higher position to further increase the photographing area of the camera. However, this method is not suitable for a camera which is located at a distance from the focal distance There arises a problem that distortion of the subject becomes serious.

An object of the present invention is to provide an array camera configured to photograph each of roads connected to an intersection S and an intersection S, The present invention provides an intersection photographing system using an array camera capable of acquiring an image of an intersection as well as an intersection.

In addition, another object of the present invention is to provide a method and an apparatus for measuring the size of an image of an object when images of the same size are photographed by a camera, And an image of the same image quality is acquired by allowing the zoom lens to be installed in the camera.

Further, another object of the present invention is to provide a method of correcting a perspective image generated by images captured by an array camera and correcting a perspective image into a plane image and then synthesizing the image, thereby reducing a distortion phenomenon caused by a photographing angle of the camera and a subject, The present invention provides an intersection photographing system using an array camera.

An array camera provided at one side of the intersection S and including a plurality of cameras for photographing each of the intersections S and the associated roads connected to the intersection S; An array-type camera, comprising: a controller for synthesizing photographed images obtained by an array-type camera through a panoramic technique, the system comprising: an array-type camera which, when an object of the same size is placed in focus of each of the cameras, Each of the cameras is provided with a lens having a zoom magnification proportional to its own focal length so that the size of the image of the object forms the same size.

Further, in the present invention, it is preferable that the zoom camera of the camera that photographs the linked roads located farther from the camera than the cameras that photograph the linked roads located at a close distance from the installed position of the arrayed camera is larger.

The controller may further include: an image analyzer for analyzing each of the images captured by the array camera to detect a vehicle image and a road image from the images; The vehicle image and the road image detected by the image analyzing unit are corrected in parallel to the X axis after matching each of the photographed images on a coordinate having a road direction as an X axis and a road width direction as a Y axis And the image correction unit may be further included to correct the photographed images into a planar image.

In addition, in the present invention, it is preferable that the array camera is provided with the cameras on one mounting table.

According to the present invention having the above-described objects and solutions, an array camera includes a plurality of cameras for photographing each of the link roads connected to the intersection S and the intersection S, Therefore, it is possible to monitor not only the traffic situation of the intersection but also the congestion situation of the connected roads.

According to the present invention, when an object of the same size is photographed at the focal length of each of the cameras constituting the array camera, a zoom lens that is proportional to the focal length of the camera, It is possible to increase the monitoring efficiency through the image by making it possible to accurately pick up an object at a distance.

In addition, according to the present invention, a perspective image in an image generated due to a photographing angle of a camera and a subject is corrected to a plane image, thereby reducing a distortion of an image, thereby further enhancing the monitoring efficiency.

1 is an actual photograph showing a conventional intersection image.
2 is an actual image of an intersection disclosed in Korean Patent Laid-Open No. 10-2011-0105527 (entitled " Integrated Traffic Light Management Apparatus and Method Including an Intersection Accident Recording and Imaging Apparatus ").
Fig. 3 is a configuration diagram for explaining the overall configuration of the present invention.
Fig. 4 is an actual screen showing intersections and connection roads taken by the array camera of Fig. 3; Fig.
Fig. 5 is an exemplary view for explaining the array camera of Fig. 2. Fig.
Fig. 6 is a block diagram showing the controller of Fig. 3; Fig.
7 is a block diagram showing the operation of the image correction unit of FIG.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

Fig. 3 is a configuration diagram for explaining the overall configuration of the present invention, and Fig. 4 is an actual photograph showing intersections and connection roads taken by the array camera of Fig.

The intersection photographing system 1, which is an embodiment of the present invention, photographs each road (hereinafter, referred to as a linkage road) connected to an intersection S and an intersection S by being installed at a side of an intersection, A controller 5 for analyzing the photographed images transmitted from the array camera 3 to correct distortion of each of the images and then combining the images, And a data communication network 8 for providing a data movement path between the controller 5 and the control center 7.

Fig. 5 is an exemplary view for explaining the array camera of Fig. 2. Fig.

The array camera 3 includes five cameras 3-1, 3-2, 3-3, 3-4, and 3-5 as shown in Fig. 5, At this time, the central camera 3-5 connects the intersection S to the first camera 3-1 and the second camera 3-2 from the first camera 3-1 and the second camera 3-2, The third camera 3-3 and the fourth camera 3-4 photograph the third linkage road S3 and the fourth linkage road S4 which are distant from the position of the third camera 3-3 and the fourth camera 3-4.

The first camera 3-1 and the second camera 3-2 are set to focus on the first linkage road S1 and the second linkage road S2 which are adjacent roads from their positions. At this time, the first camera 3-1 and the second camera 3-2 include a low magnification lens so as to capture a near vision image.

The third camera 3-3 and the fourth camera 3-4 are set to focus on the third linkage road S3 and the fourth linkage road S4 which are roads remotely located from their positions. At this time, the third camera 3-3 and the fourth camera 3-4 include a high magnification lens to capture a long distance image, thereby acquiring a long distance image of the same image quality.

The central camera 3-5 acquires an intersection image by photographing the intersection S, and in detail, it is preferable that the central camera 3-5 is a fish-eye lens which is an ultra-wide angle lens having a square angle exceeding 180 degrees .

The central camera 3-5 and the first, second, third, and fourth cameras 3-1, 3-2, 3-3, (S1), (S2), (S3), and (S4) based on the center of each of the photographing areas S, S1, S2, S3, and S4.

At this time, the first, second, third, and fourth cameras 3-1, 3-2, 3-3, 3-4, A zoom lens that is proportional to its own focal length is installed so that images of the object in the photographed image are formed to have the same size.

Accordingly, when a vehicle of the same size is photographed at a long distance and at a close distance, it is photographed at a size smaller than that of a vehicle image in a near-distance shooting image rather than a vehicle image in a remote shooting image, thereby solving the problem that accurate traffic situation information can not be monitored through a remote shooting image .

FIG. 6 is a block diagram showing the controller of FIG. 3, and FIG. 7 is a diagram showing the operation of the image correction unit of FIG.

The controller 5 of Fig. 6 includes a database unit 52 for storing data, a data transmission / reception unit 53 for transmitting / receiving data to / from the control center 7, a camera 3-1 of the array- A camera driver 55 for driving the cameras 3-1, 3-2, 3-3, 3-4, 3-5, and a camera driver 55 S3 and S4 of the intersection S and the first, second, third, and fourth linkage roads S1, S2, S3, and S4, An image correcting unit 57 for correcting the vehicle image and the road image detected by the image analyzing unit 56 into a plane image, an image correcting unit 57, An image synthesizing unit 59 for combining the images corrected by the control unit 52 into one image and a control unit 52 for controlling the control objects 52, 53, 55, 56, 57, (51).

The control unit 51 manages and controls the control objects 52, 53, 55, 56, 57,

In addition, the control unit 51 periodically transmits the combined images stored in the database unit 52 to the control center 7.

In addition, the control unit 51 inputs the acquired image to the image analysis unit 56 when the image is acquired under the control of the camera driving unit 55.

When the vehicle image and the road image are detected by the image analysis unit 56, the control unit 51 inputs the detected information and the captured image to the image correction unit 57

When the correction of each image is completed by the image correction unit 57, the control unit 51 inputs the corrected images to the image synthesis unit 59.

The control unit 51 also stores the composite image combined by the image composition unit 59 in the database unit 52 and then periodically controls the data transmission and reception unit 53 to transmit the stored composite image to the control center 7 do.

The camera driving unit 55 controls the above-described array type camera 3 of Fig. 4, and the array type camera 3 is connected to the connecting roads S1, S2, S3, S4, (3-3), (3-4), (3-5) for photographing the respective sections of the video camera 10-1, 10-2, and 10-3.

At this time, the cameras 3-1, 3-2, 3-3, 3-4, and 3-5 are installed on the upper part of one installation, A zoom lens that is proportional to its own focal length is installed so that the image of the object is captured in the same size from the obtained image.

The camera driving unit 55 also inputs the captured image obtained by the shooting of the array camera 3 to the image analysis unit 56 under the control of the control unit 51. [

The image analyzing unit 56 analyzes the inputted shot image by using a predetermined image analysis algorithm, and detects a road image representing a road in the shot image and a vehicle image representing a vehicle.

The image analyzer 56 also inputs information about the detected road image and the vehicle image and the captured image to the image corrector 57.

The image correcting unit 57 corrects the perspective sensation generated from the images of the cameras 3-1, 3-2, 3-3, 3-4, and 3-5 into a plane image, Specifically, the vehicle image and the road image are detected from the photographed image by using the detection information of the vehicle image and the road image input from the image analysis unit 56, and then the detected vehicle image and road image are corrected to the plane image.

7, the photographed image 300 is placed on the coordinate system consisting of the X axis representing the road direction and the Y axis representing the road width direction, and the image 300 of the array camera 3 It can be seen that the length of the road width d (Y value) decreases as the mounting position P is located on the left side on the coordinate (farther from the position P of the camera 3 have.

Accordingly, as the X value of the vehicle image 303 in the photographed image 300 also increases, the width (Y value) of the vehicle is reduced. That is, in the actual scene, although the longitudinal direction of the vehicle is horizontal to the X axis (road direction), the longitudinal direction of the vehicle image 303 representing the vehicle is formed to be inclined from the X axis (road direction) .

The image corrector 57 corrects (303 ') the vehicle image 303 of the photographed image 300 formed as described above horizontally to the X axis (road direction), and the road image 301 representing the road is also corrected to the X axis And corrects the photographed image 300 as a plane image 300 'by correcting the image 300' horizontally.

After the correction of each image is performed, the image correction unit 57 inputs the corrected images to the image synthesis unit 59 under the control of the control unit 51. [

The image synthesizer 59 generates a panorama image obtained by combining the images input from the image corrector 57 with the images corrected in correspondence with the positions in the actual field (space) using the panoramic technique.

The panorama image generated by the image composition unit 59 is stored in the database unit 52. [

As described above, the intersection photographing system 1, which is an embodiment of the present invention, includes the cameras 3-1, 3-2, 3-3, 3-4, 3-5 are arranged to photograph the intersection S and each of the linkage roads S1, S2, S3 and S4 so that the linkage roads S1, S2, (S3) and (S4), and thus it is possible to monitor the traffic situation of an intersection at an accurate intersection.

In addition, the intersection photographing system 1 is configured such that the cameras 3-1, 3-2, 3-3, 3-4, and 3-5 photograph the middle of each photographing section by focusing The zoom lens is installed in each camera so that the size of the object image is the same as the size of the object image when the object having the same size is located at each of the focal points and is photographed so that the object at a distance can be accurately monitored do.

In addition, the intersection photographing system 1 can recognize the perspective feeling that occurs in each image obtained by the cameras 3-1, 3-2, 3-3, 3-4, and 3-5 By correcting the image with the plane image, the distortion caused by the camera and the photographing angle of the subject can be remarkably reduced and the monitoring efficiency can be improved.

1: Intersection shooting system 3: Array type camera 5: Controller
7: control center 8: data communication network 51:
52: Database part 53: Data transmission / reception part 55: Camera driving part
56: image analysis unit 57: image correction unit 59: image synthesis unit

Claims (4)

An array camera provided at one side of the intersection S and configured to photograph each of the intersection S and the associated roads connected to the intersection S; And a controller that synthesizes the panoramic image by a panoramic technique, the system comprising:
The array-
A camera having a zoom magnification proportional to its own focal length is installed in each of the cameras so that the size of the image of the object in the captured image is the same when the same size object is located in the focus of each of the cameras And the intersection photographing system. 2. The intersection photographing system according to claim 1, wherein the camera for photographing the link roads located farther away from the camera for photographing the link roads located near from the installation position is larger in magnification. The system of claim 2, wherein the controller
An image analyzer for analyzing each of the photographed images obtained by the array camera to detect a vehicle image and a road image from each photographed image;
The vehicle image and the road image detected by the image analyzing unit are corrected in parallel to the X axis after matching each of the photographed images on a coordinate having a road direction as an X axis and a road width direction as a Y axis And the image correction unit corrects the photographed images into a planar image. The intersection photographing system according to any one of claims 1 to 3, wherein the array camera includes the cameras installed on one mount.

Images