KR101630596B1 - Photographing apparatus for bottom of car and operating method thereof - Google Patents

Abstract

The present invention relates to a device for photographing the bottom of a vehicle and to an operating method thereof. The device for photographing the bottom of a vehicle comprises: a camera for photographing the image of the bottom surface of a vehicle; a lighting module for radiating light to the bottom surface of a vehicle to be photographed by the camera; and a correction module for correcting the image of the bottom surface of the vehicle obtained by the camera. The correction module includes: a perspective distortion correcting part for correcting perspective distortion generated by a difference in the distance of the bottom surface of the vehicle from the camera depending on the view angle of the camera; and a matching part for specifying a corresponding point with corresponding RGB values by comparing the RGB values of each pixel of the images corrected by the perspective distortion correcting part, and matching the multiple images into a single image with respect to the corresponding point. The device for photographing the bottom of a vehicle according to the present invention can accurately detect a dangerous object or the like, which can be mounted on the bottom surface of a vehicle, by photographing the bottom surface without perspective distortion.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a vehicle bottom photographing apparatus,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle underground photographing apparatus and a vehicle underground photographing method for operating the same, The present invention relates to a vehicle bottom photographing apparatus that matches an image with a clear one and a vehicle bottom photographing method that operates the same.

Vehicles are an easy way to travel. Threats of terrorism using mobile means are not being cut off from all parts of the world. Especially in the case of underground vehicles, it is not easy to monitor and is used as a tool of terrorism frequently to terrorists. In order to monitor the floor of the vehicle, it is most common to use the reflection mirror to visually check the bottom of the vehicle. However, this is only the level of the edge of the floor of the vehicle and the center of the floor of the vehicle is difficult to see. Due to its nature, it is very difficult to identify signs of dangerous substances.

Korean Patent Laid-Open Publication No. 10-2011-0043048 is composed of a lower illuminator illuminating a lower portion of a moving vehicle and a lower inspector having a camera mounted on a lower portion of the vehicle to photograph a bottom surface of the vehicle. Korean Patent Laid-Open Publication No. 10-2011-0043148 improves the dark environment of the vehicle floor when shooting through the lower lighting. However, due to the nature of the vehicle, the angle of view of the camera is generally wide The distance is shortened. When the angle of view is widened, the center of the vehicle's bottom surface facing the center of the camera can be displayed as the real size, but the outer portion of the vehicle's bottom surface shows a pronounced perspective distortion. When this kind of perspective distortion is severe, it is difficult to detect dangerous objects mounted on the vehicle because the image of the vehicle floor is distorted.

Korean Patent Laid-Open Publication No. 10-2014-0043148 discloses a control calculation device that calculates the speed of a moving vehicle, captures images of a plurality of vehicle floor surfaces, and forms them into one image. Such an image matching method is a method that depends only on the moving speed of the vehicle, and the occurrence of errors frequently occurs in image matching. For example, the speed at which the vehicle passes over the lower inspection section where the camera is disposed is not always uniform, and the vehicle speed during the time during which the vehicle is moved by the acceleration force at the stop position has a large difference from the average speed. Also, it is difficult for the speed of the moving vehicle to always keep the same speed uniformly. Therefore, matching of plural images in this manner is not accurate and detection of dangerous substances can not be performed properly.

Korean Patent Laid-Open Publication No. 10-2014-0043148 (vehicle under photographing apparatus)

As a result of various studies to solve the above problem, the inventors of the present invention have found that a distance distortion correction unit that corrects a perspective distortion of a photographed image by calculating a difference between a viewing angle of the camera and a vehicle floor from the camera, It is possible to photograph a clear bottom surface of a vehicle without distortion by means of a vehicle under photographing apparatus including an image matching unit for specifying a corresponding point having a same RGB (Red-Green-Blue)

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a vehicle underground photographing apparatus and a photographing method for operating the same, which can acquire a clear bottom image of a vehicle with no distortion of perspective and a high matching state.

In order to achieve the above object, a vehicle lower imaging apparatus according to an embodiment of the present invention includes:

A camera for photographing an image of a vehicle floor;

An illumination module for illuminating the bottom surface of the vehicle to be photographed by the camera; And

And a correction module for correcting the image of the vehicle floor obtained by the camera,

The correction module

A perspective distortion correcting unit for correcting a perspective distortion generated by a difference in distance between the camera and a bottom surface of the vehicle according to an angle of view of the camera; And

And an image matching unit for comparing the RGB values per pixel of the images corrected by the perspective distortion correcting unit to specify corresponding points where the RGB values coincide with each other and matching the plurality of images to one image based on the corresponding points .

To induce multiple images to be more precisely matched,

The image matching unit may further determine a corresponding point by additionally comparing transparency (alpha) values to RGB values per pixel to detect corresponding points of the plurality of images.

Further, in order to induce precise image registration in which the transparency (alpha) value is reflected,

The lighting module

A light source unit for irradiating light on the vehicle floor;

A sensor unit for receiving light reflected from a vehicle floor; And

And a light source control unit for controlling the light source so that the entire bottom surface of the vehicle has a constant illuminance by comparing the illuminance of the light received from the sensor unit.

The present invention provides a vehicle lower photographing apparatus and a method of operating the same that corrects a perspective distortion of an image generated when a camera is positioned close to a vehicle bottom when photographing a bottom surface of the vehicle, The plurality of images are matched after specifying the corresponding points of the plurality of images by comparison, so that it is possible to provide an image of a clear bottom surface of the vehicle free from distortion. Therefore, the vehicle underground photographing apparatus and the method of operating the same according to the present invention can accurately detect various dangerous materials that can be installed on the floor of the vehicle.

1 is a plan view of an under-vehicle photographing apparatus according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a correction module for operating the vehicle lower imaging apparatus shown in Fig. 1. Fig.
3 is an image obtained by correcting the perspective distortion and the perspective distortion generated when the vehicle is taken close-up.
FIG. 4 is a schematic diagram for explaining a process of matching the plurality of images by the image matching unit shown in FIG. 2. FIG.
5 is a perspective view of the vehicle lower imaging system on which the vehicle lower imaging apparatus shown in Fig. 1 is mounted.

The following detailed description of the invention refers to the accompanying drawings, which illustrate, by way of illustration, specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. It should be understood that the various embodiments of the present invention are different, but need not be mutually exclusive. For example, the specific shapes, structures, and characteristics described in this specification may be modified and changed from one embodiment to another without departing from the spirit and scope of the present invention. It is also to be understood that the position or arrangement of the individual components in each embodiment may be varied without departing from the spirit and scope of the present invention. Therefore, the following detailed description is not to be taken in a limiting sense, and the scope of the present invention should be construed as encompassing the scope of the appended claims and all equivalents thereof. In the drawings, like reference numbers designate the same or similar components throughout the several views.

Hereinafter, various embodiments of the vehicle lower imaging apparatus of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a plan view of a vehicle lower imaging apparatus according to an embodiment of the present invention, and FIG. 2 is a schematic diagram of a correction module for operating the vehicle lower imaging apparatus shown in FIG. 1 and 2, the vehicle underground photographing apparatus 100 of the present invention photographs the bottom surface of the vehicle while moving the lower portion of the vehicle bottom surface. The vehicle underground photographing apparatus 100 includes a camera 110 for photographing an image of the vehicle floor, a lighting module 120 and a correction module 130 (not shown in Fig. 1) for correcting the photographed image.

The vehicle lower photographing apparatus 100 photographs the bottom surface of the vehicle while moving the lower portion of the vehicle bottom surface. The camera 110 may have a single camera or a plurality of cameras so that the camera 110 can be wider than the bottom of the vehicle. For example, the camera 110 may include a center camera 111 and an edge camera 112, 113. The center camera 111 may be disposed at a position parallel to the bottom surface of the vehicle so as to photograph the center of the bottom surface of the vehicle. The edge cameras 112 and 113 are disposed obliquely to the vehicle floor or the center camera 111 at an oblique angle, and can photograph the edges of the vehicle floor. As another example, the center camera 111 and the edge cameras 112 and 113 may be spaced apart from each other at positions parallel to the vehicle floor.

The center camera 111 and the edge cameras 112 and 113 are not particularly limited as long as the cameras can be used to acquire images. For example, the center camera 111 and the edge cameras 112 and 113 are CCD (Charge-coupled device ) Camera. 3- (1), the photographing apparatus 100 mounted with the camera performs photographing at a position close to the position of the bottom surface of the vehicle due to the characteristics of the vehicle, And the bottom surface of the vehicle is photographed under a condition that the angle of view of the vehicle is widened. The image of the floor of the vehicle photographed in such an environment causes severe distortion of the perspective distortion as shown in Fig. 3- (2).

3- (3) is an image of the bottom surface of the vehicle obtained by correcting the perspective distortion. 3- (3) and Fig. 3- (2), the center of the vehicle floor facing the camera does not have a perspective distortion, but the distance from the edge of the vehicle Distortion occurs. The distance Z from the lens position of the center camera 111 to the edge of the vehicle floor surface is smaller than the distance Z from the lens position of the center camera 111 to the edge of the vehicle floor surface as shown in Figure 3- (2) The distance to the center becomes longer than the distance A, so that a perspective distortion occurs.

The distortion of the perspective distortion narrowly displays the width of the edge of the vehicle floor surface, thereby making it difficult to check the dangerous substances installed on the vehicle floor.

Referring to FIGS. 1 and 2 again, the correction module 130 corrects the perspective distortion shape of the image obtained by the center camera 111 and the edge cameras 112 and 113, The images are matched and provided as one image. The correction module 130 may include a perspective distortion correction unit 131, an image matching unit 132, and a distance measurement unit 133.

The perspective distortion correction unit 131 corrects the perspective distorted image obtained from the camera 110. [ For example, the distance measuring unit 133 may measure the distance between the position of the camera 110 and the floor of the vehicle. The distance measuring unit 133 can calculate the distance from the camera 110 to the vehicle floor surface by, for example, irradiating infrared rays and measuring the time of the infrared ray reflected on the vehicle floor surface, 133 may generate ultrasonic waves to obtain the time of the ultrasonic waves that are reflected on the floor of the vehicle and then return to calculate the distance from the camera 110 to the floor of the vehicle. The distance measuring unit 133 measures the distance between the camera 110 and the vehicle floor and sends distance information to the perspective distortion correction unit 131.

The perspective distortion correction unit 131 calculates the distance between the camera 110 and the vehicle bottom surface (A in FIG. 3- (2)) and the angle of view of the camera (FIG. 3- (2) Of B] is used to correct the distorted image. At this time, the perspective distortion correction unit 131 can acquire the information of the angle of view at the time of photographing from the camera 110.

A method of correcting the distorted image by the perspective distortion correction unit 131 will be described in detail. The image enlargement value is calculated to enlarge the image narrowed by the perspective distortion. 3 (2), the center camera 111 is disposed at a position parallel to the vehicle floor to photograph the center portion of the vehicle floor, and the edge cameras 112 and 113 are disposed at an angle So that the edge of the vehicle floor can be photographed. First, the image enlargement value G of the center camera 111 can be calculated by the following equation (1).

[Equation 1]

G = A * tan (B / 2)

[G: image enlargement value, A: distance between the center camera and the vehicle floor, B: angle of view]

Then, the perspective distortion correction unit 131 enlarges the width of the distorted image using the obtained image enlargement value G. When the perspective-distorted image is enlarged, the enlarged pixel can not be displayed until the RGB information is filled. A pixel requiring the RGB information enlarged by the image enlargement value (G) The same information as the value can be input and displayed as a color.

On the other hand, the perspective distortion correction unit 131 may calculate the image enlargement value G1 of the edge cameras 112 and 113 in the following order. First, the distance F between the edge cameras 112 and 113 and the vehicle floor is measured through a distance measuring unit 133 disposed at a position close to the edge cameras 112 and 113. The perspective distortion correction unit 131 may calculate the image enlargement value G1 of the edge cameras 112 and 113 by substituting the F distance and the A and G values of Equation 1 into Equation 2 below.

&Quot; (2) "

G1 = F * G / A

[F: Distance between the edge camera and the vehicle floor, G: Image enlargement value of the center camera, A: Distance between the center camera and the vehicle floor]

Equation 2 is obtained by using the ratio of the distance A between the center camera 111 and the vehicle floor surface to the distance between the edge cameras 112 and 113 and the vehicle floor surface, The image enlargement value G1 is relatively calculated. Then, the perspective distortion correction unit 131 enlarges the width of the distorted image obtained through the edge cameras 112 and 113 using the obtained image enlargement value G1. Pixels that are enlarged by the image enlargement value G1 and require RGB information can be displayed as colors by inputting the same information as the RGB values of the adjacent pixels.

3 (2), the edge cameras 112 and 113 may be disposed apart from each other in a horizontal plane with the vehicle floor, like the center camera 111. In this case, The image enlargement values of the edge camera 111 and the edge cameras 112 and 113 can be obtained through Equation (1) described above.

When the camera 110 photographs the bottom surface of a vehicle and sends out a plurality of distantly distorted images, the perspective distortion correction unit 131 can correct the distorted images sequentially over a plurality of images sequentially. The perspective distortion correction unit 131 sends the corrected images to the image matching unit 132. [

The image matching unit 132 matches the plurality of images received from the perspective distortion correction unit 131 so as to be displayed as one image. In order to precisely match multiple images taken on different vehicle floor surfaces, the image matching unit 132 compares the RGB values per pixel of the plurality of images to specify corresponding points where the RGB values coincide, and then overlaps the corresponding points, ≪ / RTI >

FIG. 4 is a schematic diagram for explaining a process of matching the plurality of images by the image matching unit shown in FIG. 2. FIG. Referring to Figs. 2 and 4, the image matching unit 132 may screen pixels having the same RGB value of images to match a plurality of images. When a plurality of images are enlarged, a large number of pixels are combined to form an image. Each of the pixels has a unique RGB value (pixel information). As shown in FIG. 4, the image matching unit 132 specifies pixels corresponding to RGB values as corresponding points (Y), overlaps a plurality of images based on the corresponding points (Y) and provides them as one image data do.

Here, the correspondence point (Y) can be set to a plurality of pixels having the same RGB value. As illustrated in Fig. 4, the image matching unit 132 may set these ten pixels as the corresponding points when the ten or more pixels successively arranged on the image have the same RGB value. 4, the number of pixels that can be specified by the corresponding points can be variously changed in consideration of the processing speed of the image matching unit 132 and the accuracy of matching.

Preferably, in order to detect the corresponding point Y of the plurality of images, the image registration unit 132 may additionally compare the transparency (alpha) value to the RGB value per pixel to specify the corresponding point. Transparency (α) is information about turbidity which is optically measured by chromaticity and scattering degree of light which ignores brightness. By adding the transparency value to the RGB value and specifying the corresponding points of the plurality of images, the accuracy of matching can be improved.

The image matching unit 132 can send an image of the matched vehicle floor surface to the display device and the inspector can inspect the vehicle floor surface through the display device.

On the other hand, since the image matching unit 132 specifies the corresponding point based on the RGB values and the transparency, the illuminance of the illumination illuminating the lower portion of the vehicle must be maintained constant. If the illuminance of the illumination is not constant, the RGB value and the transparency of the vehicle floor surface are changed, and it is difficult for the image matching unit 132 to specify the corresponding point, and an error that can not be matched can occur. Especially, when the transparency is used as a criterion for specifying the corresponding point, the uniformity of illuminance illuminating the lower portion of the vehicle is an important factor.

Referring to FIGS. 1 and 2, the lighting module illuminates the bottom surface of a vehicle to be photographed by the camera. The illumination module 120 includes a light source unit 121, a sensor unit 122, and a light source control unit 123.

The light source unit 121 illuminates the bottom surface of the vehicle and brightens the dark shooting environment due to the characteristics of the lower portion of the vehicle. The light source unit 121 may include a plurality of LED chips, and the number of the LED chips may be variously changed in consideration of the size of the vehicle under photographing apparatus 100 and the vehicle to be photographed.

The sensor unit 122 receives reflected light that is reflected by the bottom surface of the vehicle and is emitted from the light source unit 121. The sensor unit 122 senses the illuminance information of the received reflected light and transmits the illuminance information to the light source control unit 123. 1, two sensor units 122 are disposed on both sides of the camera 110. However, the sensor unit 122 may be variously changed in consideration of the number of LED chips mounted on the light source unit 121 and the number of cameras.

The light source control unit 123 collects illumination information received from the sensor unit 122 and compares the illumination information with each other. The light source control unit 123 may control the light source unit 121 to increase the light irradiation amount on the LED chip in which the sensor unit 122 is located when the sensor unit 122 with low light intensity is identified in the collected lightness information have.

The light source control unit 123 controls the uneven illuminance that may be generated by the external environment or the shape characteristic of the vehicle floor surface to be uniform so that the image matching unit 132 can identify the corresponding points of the plurality of images without error do.

Hereinafter, a vehicle lower imaging system on which the vehicle lower imaging apparatus of the present invention is mounted will be described as an example.

5 is a perspective view of the vehicle lower imaging system on which the vehicle lower imaging apparatus shown in Fig. 1 is mounted. 5, the vehicle lower imaging system 1000 includes a vehicle lower imaging apparatus 100, a stage 200, and a guide unit 300. As shown in FIG.

Since the vehicle lower imaging apparatus 100 is the same as the one described above, a detailed description thereof will be omitted. A vehicle is positioned above the stage 200. The portion of the stage 200 where the wheels of the vehicle are positioned may protrude slightly higher. When viewing the stage 200 as a whole, the stage 200 may have a shape that is inclined in one direction, which allows the rainwater to drain naturally.

A laser or an infrared sensor capable of confirming whether or not the vehicle has stopped at a predetermined position of the stage 200 can be attached to the stage 200. [ In addition, the stage 200 can be mounted with a detection camera 400 for detecting the vehicle number, and can recognize the number of the vehicle entering the stage 200.

The guide unit 300 extends along the longitudinal direction of the center of the stage 200 and moves the vehicle lower imaging apparatus 100. Therefore, the vehicle lower imaging apparatus 100 can move along the guide section 300 by photographing the bottom surface of the vehicle.

Hereinafter, a vehicle bottom photographing method for operating the vehicle lower photographing apparatus of the present invention will be described.

A method for photographing a lower portion of a vehicle according to the present invention includes the steps of: (A) photographing a plurality of images of a lower portion of a vehicle using a camera; (B) (C) comparing the RGB values per pixel of the plurality of images in which the perspective distortion is corrected, and overlapping the plurality of images on the basis of corresponding points where the RGB values coincide with each other to match the images with one image .

In step (A), light is irradiated to the bottom surface of the vehicle to be photographed (A-1). Next, the light reflected from the vehicle floor is received to measure the illuminance of the bottom surface of the vehicle, and the light amount of the light source is adjusted so that the illuminance of the light reflected on the bottom surface of the vehicle is uniform (A-2). Thereafter, the camera photographs the bottom surface of the vehicle (A-3).

Thereafter, in step (B), the distance between the camera and the floor of the vehicle to be photographed is measured (B-1), and then the angle of view of the camera is checked (B-2) (B-3), and the distorted part of the perspective distorted images is enlarged to the image enlargement value (B-4). Since the method of calculating the image enlargement value has been described above, the detailed description will be omitted. Then, the pixels of the enlarged image are displayed with the same color as the RGB values of the adjacent pixels before the correction (B-5).

(C) comparing the RGB values per pixel between the calibrated images to identify corresponding points having at least 10 identical RGB values arranged continuously on the image (C-1), overlapping the plurality of images with the corresponding points Match with one image (C-2). The correspondence point may include pixels having the same RGB value and transparency (alpha) value per pixel of the plurality of images.

As described above, according to the present invention, when the bottom surface of the vehicle is photographed, the camera corrects the perspective distortion of the image generated as the camera is located close to the bottom surface of the vehicle, Since a plurality of images are matched after specifying the corresponding points of the plurality of images by comparing the transparency, it is possible to acquire an image of a clear bottom surface of the vehicle without distortion. Therefore, the vehicle underground photographing apparatus and the method of operating the same according to the present invention can accurately detect dangerous materials or the like that can be installed on the floor of the vehicle.

1000: Under Vehicle Shooting System
100: Vehicle bottom photographing apparatus
110: camera 111: center camera
112, 113: edge camera
120: illumination module 121: light source part
122: sensor unit 123: light source control unit
130: Correction module 131: Perspective distortion correction section
132: image matching unit 133: distance measuring unit
G: Image enlargement value
200: stage
300: guide portion
400: Detection camera
B: angle of view
Y: Corresponding point

Claims (13)

A camera for photographing an image of a vehicle floor;
An illumination module for illuminating the bottom surface of the vehicle to be photographed by the camera; And
And a correction module for correcting an image of a vehicle floor obtained by the camera,
The lighting module
A light source unit for irradiating light on the vehicle floor;
A sensor unit for receiving light reflected on a vehicle floor; And
And a light source control unit for comparing the illuminance of the light received from the sensor unit and controlling the light source unit so that a bottom surface of the vehicle has a constant illuminance,
The correction module
A perspective distortion correcting unit for correcting a perspective distortion generated by a difference in distance between the camera and a bottom surface of the vehicle according to an angle of view of the camera; And
A transparency (alpha) value is added to RGB values per pixel of the images corrected by the perspective distortion correction unit to specify corresponding points whose values coincide with each other, and a plurality of images are matched to one image with reference to the corresponding points And an image registration unit for outputting the image registration information. delete The method according to claim 1,
The correction module
Further comprising a distance measuring unit for measuring a distance between the camera and the vehicle floor,
Wherein the trapezoidal distortion correcting unit corrects the perspective distorted image using the distance between the camera and the vehicle floor surface and the angle of view calculated by the camera. The method of claim 3,
The correction module calculates an image enlargement value (G) by the following equation (1)
Wherein the trapezoidal distortion correcting unit enlarges the width of the perspective-distorted image based on the image enlargement value.
[Equation 1]
G = A * tan (B / 2)
[G: image enlargement value, A: distance between the camera and the vehicle floor, B: angle of view] 5. The method of claim 4,
The perspective distortion correction unit
Wherein the color of the pixel of the image enlarged by the image enlargement value is expressed by the same value as the RGB value of the adjacent pixel. The method according to claim 1,
The image matching unit
Wherein at least 10 or more pixels continuously arranged on the image are specified as corresponding points and the corresponding points of the plurality of images are matched to overlap each other. delete The method according to claim 1,
The camera
A center camera disposed at a position parallel to the vehicle floor and photographing the center of the vehicle floor; And
And an edge camera which is disposed at an oblique angle to the vehicle floor and photographs an edge of the vehicle floor surface. (A) photographing a plurality of images under the vehicle using a camera;
(B) correcting the perspective distortion of the plurality of images photographed using the angle of view of the camera and the distance between the camera and the vehicle floor; And
Vehicle bottom up containing; (C) trapezoidal distortion part compares the RGB value per pixel of the corrected multi-image by overlapping multiple image based on the corresponding points of the RGB value match (Overlap) the step of matching a single image In the method,
The step (A)
(A-1) irradiating light to a bottom surface of a vehicle to be photographed using a light source;
(A-2) receiving light reflected from a bottom surface of the vehicle to measure the illuminance of the bottom surface of the vehicle, and adjusting the light amount of the light source so that the illuminance of the light reflected on the bottom surface of the vehicle is uniform; And
(A-3) photographing the bottom surface of the vehicle by the camera,
Wherein the corresponding point in the step (C) indicates a pixel having the same RGB value and transparency (alpha) value per pixel of the plurality of images. delete 10. The method of claim 9,
The step (B)
(B-1) measuring a distance A between the camera and a bottom surface of a vehicle to be photographed;
(B-2) checking an angle of view B of the camera; And
(B-3) calculating an image enlargement value using the following equation (1);
(B-4) enlarging distorted parts of the perspective distorted images to the image enlargement value; And
(B-5) displaying the color in the pixels of the enlarged image at the same numerical value as the RGB value of the adjacent pixel before the correction.
[Equation 1]
G = A * tan (B / 2)
[G: image enlargement value, A: distance between the camera and the vehicle floor, B: angle of view] 10. The method of claim 9,
The step (C)
(C-1) comparing the RGB values per pixel between the corrected images to identify corresponding points having at least 10 identical RGB values arranged continuously on the image; And
(C-2) overlapping the plurality of images with the corresponding points to match with one image. delete

Images