JP7093962B2 - Vehicle image generator and vehicle image generator - Google Patents

Description

Application of Article 30, Paragraph 2 of the Patent Law 1. Publication date of the invention January 5, 2018 2. Place where the invention was published Photo studio in JU Gifu Hashima Auto Auction Co., Ltd. 3. Publisher name JU Gifu Hashima Auto Auction 4. Contents of the invention Use of a vehicle image generator and a vehicle image generator

The present invention relates to a vehicle image generator or a vehicle image generation program that cuts out an image within a predetermined range in which a vehicle image exists from the image data and generates another image data.

As a technique for displaying a vehicle image, for example, there is a "used car auction system" disclosed in Patent Document 1 below. In this technology, used cars that are the target of the auction are shot one by one prior to the auction. For example, for a still image, a vehicle stopped at a still image shooting location is photographed by four cameras from four directions. Then, as disclosed in FIG. 4 of the same document 1, these images are displayed on the preview information display screen of the computer. As a result, the auction participant can visually confirm the state of the target vehicle by the still image or the like in the preview information display image.

Japanese Patent Application Laid-Open No. 2003-162653

However, the vehicles targeted for auction vary in type and size. Therefore, for example, in the disclosed technique of Patent Document 1, when four cameras are fixed at predetermined positions, it is not always possible to take a vehicle image with an optimum size. Even if the fixed camera has a zoom function, it is complicated for the worker to adjust the zoom every time the camera is taken, and it is difficult to improve the work efficiency.

Such a problem can be solved, for example, by cutting out a vehicle image from image data taken by using a camera having a high resolution and aligning the vehicle image with an optimum size. However, it is not easy to detect the contour of the vehicle image in the image recognition process by the computer. Therefore, there is a problem that it may take a lot of money and time to develop the system.

The present invention has been made to solve the above-mentioned problems, and is a vehicle image generation device capable of cutting out a vehicle image with a predetermined size and generating another image data without detecting the contour of the vehicle image. And to provide a vehicle image generation program.

In order to achieve the above object, the technical means of claim 1 or claim 3 described in the claims is adopted. According to this means, the image recognition means recognizes two circles or ellipses from the image data including the vehicle image, and the distance measuring means measures the distance between the centers of the two circles or ellipses recognized by the image recognition means. do. The frame determining means determines a predetermined frame in which a predetermined range in which the vehicle image exists can be determined based on the distance between the centers, and the image cropping means uses the predetermined frame to determine an image within a predetermined range in which the vehicle image exists. To generate another image data by cutting out. The other image data also includes image data in which the outside of the predetermined range in which the vehicle image exists is deleted.

The two circles present in the vehicle image where the sides of the vehicle appear are typically tires or wheels. Tires and wheels appear elliptical in vehicle images taken from the front or rear of the side of the vehicle. Therefore, the two ellipses present in such vehicle images are also likely to be tires or wheels. As a result, the fact that the image recognition means recognizes two circles or ellipses from the image data including the vehicle image is nothing but detecting the tires and wheels of the vehicle from the vehicle image. Further, compared to the contour of a vehicle image which tends to have a complicated shape, circles and ellipses are patterned, and the contour can be easily detected in the image recognition process by a computer. Therefore, it is possible to detect the tires and wheels of the vehicle from the vehicle image by the image recognition means.

When the two circles or ellipses are the tires or wheels of the vehicle, the distance between the centers of these two circles or ellipses measured by the distance measuring means corresponds to the wheelbase of the vehicle. Generally, the length of the vehicle (length in the front-rear direction and the total length of the vehicle) often depends on the distance between the front and rear tires (wheelbase). That is, the total length of the vehicle is roughly determined by the size of the wheelbase. Thereby, the frame determining means can determine a predetermined frame in which a predetermined range in which the vehicle image exists can be determined based on the distance between the centers of the two circles and ellipses. Therefore, by using such a predetermined frame and the image cutting means cutting out an image within a predetermined range, it becomes possible to generate an image in which a vehicle image exists as another image data.

Further, the technical means of claim 2 described in the claims is adopted. According to this means, the image recognition means recognizes two circles or ellipses within a range smaller than the entire image range of the image data. As a result, the target range of the image recognition processing by the computer is narrower than the case of recognizing such two circles or ellipses in the entire image range of the image data, so that it is necessary to recognize the two circles or ellipses. The time is shortened.

In the inventions of claim 1 and claim 3 , the tires and wheels of the vehicle can be easily detected from the vehicle image by the image recognition means, and the distance between the centers of these two circles or ellipses measured by the distance measuring means. The distance corresponds to the wheelbase of the vehicle. The frame determining means can determine a predetermined frame that can determine a predetermined range in which the vehicle image exists, based on the distance between the centers of two circles or ellipses corresponding to the wheelbase of the vehicle. Therefore, when the image cropping means cuts out an image within a predetermined range using such a predetermined frame, the vehicle image is cut out to a predetermined size without detecting the contour of the vehicle image, and another image data is generated. can do.

In the invention of claim 2, since the target range of the image recognition processing by the computer is narrower than the case of recognizing such two circles or ellipses in the entire image range of the image data, the two circles or ellipses are recognized. The time required to do this is shortened. Therefore, the image recognition process can be speeded up.

It is explanatory drawing which shows the configuration example in the case where the vehicle image photographing system which concerns on one Embodiment of this invention is applied to the photography studio for a vehicle auction. It is a figure which shows the configuration example of the camera wall which constitutes the vehicle image taking system of this embodiment. FIG. 2A is a front view, and FIG. 2B is a plan view. It is a block diagram which shows the electric structure example of the vehicle image taking system of this embodiment. It is a flowchart which shows the flow of the vehicle image taking processing executed by the computer shown in FIG. It is a flowchart which shows the flow of the subroutine "trimming frame setting process" shown in FIG. This is an example of a screen displayed on the display screen of the computer in the vehicle image capturing process shown in FIG. It is explanatory drawing which showed the conceptual image of the trimming frame setting process shown in FIG. It is explanatory drawing which showed an example of the image data before the trimming process shown in FIG. It is explanatory drawing which showed an example of the image data after the trimming process shown in FIG. It is a flowchart which showed the flow of the other reference example of the subroutine "trimming frame setting process" shown in FIG.

Hereinafter, an embodiment in which the vehicle image generation device and the vehicle image generation program of the present invention are applied to a vehicle around image capturing system will be described with reference to the drawings. This embodiment is a vehicle-around image shooting system in a studio that preliminarily shoots a vehicle body image for display on a large display installed at an auction site of a used car, a terminal device at the same venue, a terminal device connected to the Internet, or the like. It is about. Hereinafter, the vehicle-around image shooting system is also simply referred to as a “shooting system”.

<< Outline of the configuration of the shooting system 10 >>
First, the outline of the configuration of the photographing system 10 of the present embodiment will be briefly described with reference to FIG. FIG. 1 is an explanatory diagram showing a configuration example of the photographing system 10. In the coordinate system shown in the figure, the X-axis direction is the traveling direction of the vehicle 8, the Y-axis direction is the height (vehicle height) direction of the vehicle 8, and the Z-axis direction is the width (vehicle width) of the vehicle 8. The directions are shown respectively. In the present specification, such an X-axis direction may be referred to as a front-back direction. In addition, the Y-axis direction may be referred to as the vertical direction. Further, the Z-axis direction may be referred to as a left-right direction. The same applies when such a coordinate system is represented in other figures. In the present embodiment, the vehicle 8 is a medium-sized vehicle (for example, total length: 4.8 meters, total width: 1.7 meters, total height: 1.5 meters).

As shown in FIG. 1, the photography studio 2 provided with the photography system 10 is a building in which a horizontally long photography space is partitioned by a wall 3 in the traveling direction of the vehicle 8. The left side of the paper in FIG. 1 is the entrance 5, and the right side of the paper is the exit 6. The vehicle 8 as a subject stops at a predetermined position 7 when entering the shooting studio 2 from the entrance 5. The predetermined position 7 is provided substantially in the center of the shooting studio 2. Then, after the surroundings of the vehicle body and the interior of the vehicle are photographed, the vehicle exits from Exit 6. In addition, on the floor surface of the photographing studio 2, lines and marks that directly or indirectly suggest a predetermined position 7 of a rectangle long in the X-axis direction are displayed. The predetermined position 7 is set, for example, to have a long side of 6 meters and a short side of 2 meters.

The shooting system 10 provided in such a shooting studio 2 or the like is mainly composed of four camera walls 20 (20a, 20b, 20c, 20d) and a computer 11 for controlling these camera walls 20.

In the present embodiment, the four camera walls 20 are provided in the shooting studio 2, for example, the height is set to about 2.5 meters, and the length (width) in the X-axis direction is about 3 meters. Is set to. Therefore, a worker with a standard physique can hide his / her body by moving to the back side (evacuation space 4) of the camera wall 20 even in an upright posture.

In the present embodiment, these camera walls 20 are arranged at positions 2 to 3 meters away from the four corners of the predetermined position 7. That is, the camera walls 20a to 20d are provided within a predetermined distance (for example, within 3 meters) from the four corners of the vehicle 8 so as to surround the vehicle 8 stopped at the predetermined position 7 in the shooting studio 2.

For example, the camera wall 20a is provided at a position separated from the left front corner portion 7a in the exit 6 direction of the predetermined position 7 by 2 meters to 3 meters (predetermined distance). Further, the camera wall 20b is provided at a position separated from the left rear corner portion 7b in the entrance 5 direction of the predetermined position 7 by the same predetermined distance. The camera walls 20c and 20d are provided at positions separated from the right rear corner portion 7c and the right front corner portion 7d by the same predetermined distance, respectively. In FIG. 1, the alternate long and short dash line represented between the camera wall 20 (20a to 20d) and the wall 3 indicates the evacuation space 4, which will be described later.

In the shooting studio 2, the camera walls 20a to 20d arranged in this way each include three cameras 30. Further, the camera walls 20a to 20d are curved so as to surround the predetermined position 7 (the vehicle 8 stopped at the predetermined position 7), respectively. That is, the planar shapes of the camera walls 20a to 20d are formed in a curved shape along the circumference of the ellipse O drawn around the intersection (center of the predetermined position 7) P of the diagonal lines of the predetermined position 7. The ellipse O has a major axis in the X-axis direction and a minor axis in the Z-axis direction. For example, the major axis is set to 15 meters and the minor axis is set to 7 meters.

These three cameras 30 are provided so as to be arranged side by side in the peripheral direction of the predetermined position 7 (the vehicle 8 stopped at the predetermined position 7) on each of the camera walls 20. More precisely, these cameras 30 are arranged so as to be located on the circumference of the ellipse O described above. For example, the camera wall 20a provided at a position separated from the left front corner portion 7a of the predetermined position 7 by a predetermined distance is provided with cameras 30a, 30b, and 30c from the exit 6 direction toward the front left side of the vehicle 8, respectively. ing. Similarly, the camera wall 20d provided at a place separated from the right front corner portion 7d by a predetermined distance is provided with cameras 30m, 30k, and 30j from the exit 6 direction toward the front right side of the vehicle 8, respectively.

That is, these cameras 30a, 30b, 30c, 30d, 30e, 30f, 30g, 30i, 30j, 30k, 30m (hereinafter, these may be collectively referred to as "camera 30a or the like" in the drawings. Reference numeral 30 indicates that the object pointed to by the camera corresponds to any one of the cameras 30a, 30b, 30c, 30d, 30e, 30f, 30g, 30i, 30j, 30k, and 30m). It is located point-symmetrically with respect to. For example, the cameras 30a and 30g, the camera 30b and the camera 30h, the camera 30c and the camera 30i, the camera 30d and the camera 30j, the camera 30e and the camera 30k, and the camera 30f and the camera 30m are point-symmetrical with respect to the center P of the predetermined position 7, respectively. Is located in.

As a result, the cameras 30a, 30b, and 30c of the camera wall 20a can take an image of the front side 8a of the vehicle 8 and an image of the left side 8c in the front direction and the like. Further, the cameras 30m, 30k, and 30j of the camera wall 20d make it possible to take an image of the front side 8a of the vehicle 8 and an image of the right side 8d in the front direction and the like.

Further, the camera wall 20b provided at a place separated from the left rear corner portion 7b by a predetermined distance is provided with cameras 30f, 30e, and 30d from the entrance 5 direction toward the rear left side of the vehicle 8, respectively. Further, the camera wall 20c provided at a position separated from the right rear corner portion 7c by a predetermined distance is provided with cameras 30g, 30h, and 30i from the entrance 5 direction toward the rear right side of the vehicle 8, respectively.

As a result, the cameras 30f, 30e, and 30d of the camera wall 20b can take images of the rear side 8b and the left side 8c of the vehicle 8 in the rear direction and the like. Further, the cameras 30g, 30h, and 30i of the camera wall 20c make it possible to take images of the rear side 8b and the right side 8d of the vehicle 8 in the rear direction and the like.

On the other hand, the computer 11 is provided in a room (for example, an operator room or an adjustment room) separately from the shooting studio 2. In the present embodiment, the computer 11 is connected to 12 cameras 30a and the like provided on each of the camera walls 20a to 20d, and performs image processing and the like as described later. The configuration of the computer 11 and its peripheral devices will be described later with reference to FIG.

<< Configuration of camera wall 20 >>
Here, the configuration of the camera wall 20 will be described with reference to FIG. Of the four camera walls 20a to 20d, the camera wall 20a and the camera wall 20c are similarly configured. Further, the camera wall 20a and the camera wall 20b are configured to be mirror-symmetrical with respect to the YZ plane, and the camera wall 20a and the camera wall 20d are configured to be mirror-symmetrical with respect to the XY plane. Therefore, among the four camera walls 20a to 20d, the camera wall 20a will be described here as a representative. FIG. 2A shows a front view of the camera wall 20a. Further, FIG. 2B shows a plan view of the camera wall 20a. For convenience, in FIG. 2, the transparent panel 39 is colored light gray and the dome cover 38 is colored dark gray.

As shown in FIG. 2, the camera wall 20a is mainly composed of a vertical frame 21, a horizontal frame 22, a top panel 23, a side panel 24, front panels 25, 26 and the like. The vertical frame 21 and the horizontal frame 22 are, for example, aluminum L-shaped angles or square pipes, and form the framework structure of the camera wall 20a (see FIG. 2B). The horizontal frame 22 spanning in the X-axis direction is curved in a shape along the elliptical arc of the ellipse O shown in FIG. 1 or an elliptical arc close to the elliptical arc. The horizontal frames 22 are provided at a plurality of locations (for example, three locations) at different positions in the height direction, for example.

The top panel 23, the side panel 24, and the front panels 25, 26 have, for example, a wall body having a plate wall structure composed of a flat flat plate (iron plate) made of iron, and the surface thereof is painted gray. The top panel 23 plays the role of a top plate that closes the upper part of the camera wall 20a, and the side panel 24 plays the role of a side plate that closes the sides of the camera wall 20a. Since the horizontal frame 22 spanning in the X-axis direction is curved in a shape along an elliptical arc as described above, the front surface of the camera wall 20a is composed of a plurality of front panels 25 and 26.

In the present embodiment, the front panels 25 and 26 are all set to have the same length in the X-axis direction (short direction), and the total number of front panels 25 and 26 is, for example, at a predetermined position 7. The number S (= N × 2-1) obtained by doubling the number N of the cameras 30 arranged side by side in the peripheral direction and subtracting 1 is set. Since the camera wall 20a has three cameras 30, the total number S of the front panels 25 and 26 is 5 (= 3 × 2-1).

In the present embodiment, as will be described next, since the front panel 25 covers the portion to which the camera 30 is attached, the number of the three front panels 25 and the two front panels 26 is the same as the number of the cameras 30. The front surface of the camera wall 20a is configured by the above. It is not necessary to select a flat flat panel as the front panels 25 and 26, and a curved panel having a curved surface along an elliptical arc in the lateral direction may be used for the front panels 25 and 26.

The front panel 25 is divided into a plurality of walls. That is, the front panel 25 is composed of three types of wall bodies, that is, a curtain plate wall 25a, a blank plate wall 25b, and a wainscot wall 25c from above in the Y-axis direction. On the other hand, the front panel 26 is composed of a strip-shaped long wall body.

The curtain plate wall 25a is a wall body that closes the upper side of the blank plate wall 25b and the like, and the waist plate wall 25c is a wall body that closes the lower side of the blank plate wall 25b and the like. The blank plate wall 25b is a wall body that closes the front opening of the rack space that does not accommodate the camera case 33. In the present embodiment, the camera case 33 is housed in the middle rack space 28, and the camera case 33 is not housed in the upper rack space 27 and the lower rack space 29. Therefore, in the present embodiment, the blank plate wall 25b closes the upper and lower rack spaces 27 and 29. The front lid 36 of the camera case 33 is attached to the rack space 28 in the middle stage. In FIG. 2A, the rack space is represented by a two-dot chain line, and some of them are not marked.

The front panel 26 is a wall body that closes the front opening of the portion where the rack spaces 27, 28, and 29 are not provided. In the present embodiment, since the camera wall 20a is provided so that the three cameras 30 are arranged in the peripheral direction of the vehicle 8, the front panels 26 are attached to two places between the three cameras. If rack spaces 27, 28, 29 and the like are provided in the internal space of the camera wall 20a blocked by the front panel 26, the camera 30 can be attached. In such a case, a front panel 25 (curtain board wall 25a, blank board wall 25b, wainscot wall 25c) or the like is attached instead of the front panel 26.

In this embodiment, a back panel (not shown) is attached to the back of the camera wall 20a. This prevents a part of the body from entering the camera wall 20a and dust or the like from entering the camera wall 20a when the worker hides behind the camera wall 20a, as will be described later. ing. If this is not expected, the back panel may not be provided.

A camera case 33, a front lid 36, and the like are attached to the rack spaces 27, 28, and 29 in which the camera 30 is housed. In the present embodiment, the camera 30 is housed in the rack space 28 in the middle stage, and a camera box 37 having a front lid 36, a dome cover 38, and a transparent panel 39 is attached to the front opening thereof.

The camera 30 is composed of a camera body and a lens unit, which are attached to a pan head. The camera body, lens unit, and pan head are not shown. The camera body is, for example, a high-definition high-definition color camera capable of shooting a moving image, and is configured to be able to output the shot moving image data via an SDI (Serial Digital Interface). The lens unit is also equipped with an electric zoom lens mechanism, and zoom adjustment by autofocus is also possible. The zoom, white balance, exposure, etc. can be adjusted from the computer 11.

In the present embodiment, the setting data such as zoom and white balance can be received from the computer 11 connected by the RS-232C cable 17 via the camera interface 14 (see FIG. 3).

The cloud stand changes the shooting direction of the camera 30 attached to the cloud stand to the left-right direction (front-back direction of the vehicle 8) or up-down direction, or changes the position of the camera 30 to the front-back direction (left-right direction of the vehicle 8). It can be. The pan head is mainly composed of a pan / tilt unit and a slide unit. The pan / tilt unit and the slide unit are not shown.

By attaching a camera 30a or the like to each of the pan heads 40, the pan or tilt of each pan head 40 is set so that the shooting direction of each camera 30a or the like faces the center P (see FIG. 1) of the predetermined position 7. It will be adjusted. For example, when the vehicle 8 is photographed by the reference camera described later, the pan and tilt of the pan head 40 of the reference camera are adjusted so that the entire vehicle 8 is located at the center of the shooting field of view and the angle of view (shooting range). Will be done. For cameras other than the reference camera, the pan of the pan head 40 is adjusted so that the entire vehicle 8 is located at the center of the shooting field of view and the angle of view (shooting range). On the other hand, for tilt, the height of the top of the roof contour of the vehicle 8 in the captured image is the height of the top of the contour taken by the reference camera (height of the alternate long and short dash line shown in FIG. 8 (J)) Yh. Is adjusted to be the same as. That is, the height of the top of the roof contour of the vehicle 8 taken by the camera 30a or the like is the respective images 50a, 50b, 50c, 50d, 50e, 50f, 50g, 50h, 50i, 50j, 50k, 50m (hereinafter, these). May be collectively referred to as "image 50a, etc.").

As a result, the images 60a, 60b, 60c, 60d, 60e, 60f, 60g, 60h, 60i, 60j, 60k, 60m (hereinafter referred to as these) similarly trimmed in the predetermined trimming frame in the trimming process (S117) described later. The height of the roof of the vehicle 8 (the height of the alternate long and short dash line shown in FIG. 9) Yi reflected in the image (sometimes collectively referred to as “image 60a, etc.”) is also included. Therefore, by connecting the images 60a and the like having the same roof height Yi of the vehicle 8 so as to be arranged in the order of the peripheral direction of the vehicle 8, the vehicle body image of the entire circumference of the vehicle 8 can be viewed around. It can be made to appear as if it is rotated like a 360-degree image).

≪Computer 11, peripheral device configuration≫
Next, the configuration of the computer 11 and its peripheral devices will be described with reference to FIG. FIG. 3 shows a block diagram showing an example of an electrical configuration of the photographing system 10. The computer 11 is typically a personal computer, and may be, for example, an MPU, a main memory (DRAM, etc.), a storage device (magnetic disk, flash memory, etc.), a communication interface (LAN interface, SDI interface, etc.), a display, a keyboard, or the like. It is configured (these are not shown in FIG. 3).

As will be described later, a vehicle image capture processing program, a trimming frame setting processing program, and the like executed by the computer 11 are stored in advance in the storage device. By this vehicle image capturing process, the computer 11 acquires moving image data from the cameras 30a and the like of the camera walls 20a to 20d via the video hub 12 and the SDI cables 15 and 16 by the SDI interface. The computer 11 is connected to the LAN cable 18 by the LAN interface, and is also connected to the camera interface 14 of the camera 30a or the like via the RS-232C cable 17. The video signal output from the computer 11 to the display is also sent to the sub-display shown in the figure via the distributor. This sub-display is arranged in, for example, the evacuation space 4 of any of the camera walls 20a to 20d.

The video hub 12 is a concentrator that can select a plurality of channels of video signals by an SDI interface, and includes a LAN interface. In the present embodiment, the video hub 12 is configured so that, for example, a 4-channel output system can be selected for a 12-channel input system. The number of channels of the input system is matched to the number 12 of the cameras 30a and the like, and the number of channels of the output system is matched to the number of camera walls 20a to 20d. As a result, for example, each camera wall 20a to 20d can be divided into 4 channels (A to D channels) and transmitted to the computer 11. The video hub 12 is also connected to the LAN cable 18. Therefore, the transmission of the moving image data of the A to D channels is performed according to the control command received from the computer 11 via the LAN.

For example, at the transmission timing of the A channel, the video hub 12 time-divisions the moving image data captured by the cameras 30a, 30b, and 30c of the camera wall 20a into predetermined time slots and transmits the video data to the computer 11. Further, at the transmission timing of the B channel, the moving image data captured by the cameras 30d, 30e, and 30f of the camera wall 20b are time-divisioned into predetermined time slots and transmitted to the computer 11. Similarly, the moving image data of the cameras 30g, 30h, 30i of the camera wall 20c is transmitted by the C channel, and the moving image data of the cameras 30j, 30k, 30m of the camera wall 20d is transmitted by the D channel at the respective timings.

In the present embodiment, the camera interface 14 included in the cameras 30a and the like is an interface compliant with RS-232C, and the cameras 30a and the like are electrically connected in a daisy chain. When a control command with an address is received from the computer 11, the camera 30 at the address performs control according to the control command.

The number of channels included in the video hub 12 may be as long as the number of input systems is equal to or greater than the number of cameras 30a or the like, and the number of output systems may be larger, but may be 1 channel or more and less than or equal to the number of cameras 30a or the like. Further, in the present embodiment, the file server 13 and the like are connected to the LAN cable 18. The file server 13 is executed by the computer 11 and the unique information (exhibition number, vehicle type, model, exhaust volume, mileage, vehicle name, etc.) regarding the vehicle 8 described in the exhibition inspection slip, and as will be described later. Still image data and the like cut out (captured) from the moving image data by the vehicle image shooting process are accumulated. The file server is a concept including a database server.

In addition, as peripheral devices (not shown), for example, another personal computer, a printer, a wireless LAN access point, or the like is connected to the LAN cable 18. The interior image data taken by the digital camera for taking an interior image carried by the worker is stored in the file server 13 via the access point of the wireless LAN. Further, in some cases, a display device (monitor device) for constantly displaying moving image data sent from a specific camera 30 (for example, the camera 30d) of the cameras 30a or the like or a separately provided camera is provided. In this case, in the moving image data acquisition process (S107) of the vehicle image capturing process described below, it is not necessary to display the moving image on the display of the computer 11.

≪Vehicle image shooting process by computer 11≫
Subsequently, the vehicle image capturing process executed by the computer 11 will be described with reference to FIGS. 4 to 9. FIG. 4 shows a flowchart showing the flow of the vehicle image capturing process executed by the computer 11. FIG. 5 shows a flowchart showing the flow of the subroutine “trimming frame setting process” shown in FIG. FIG. 6 illustrates a screen example displayed on the display screen of the computer 11 in the vehicle image capturing process.

The software button displayed on the display screen 70 of FIG. 6 can be selected by the operator of the computer 11 by pressing the key when the corresponding key switch is present on the keyboard, but the operator is the mouse. It can also be selected by clicking the software button with a pointing device such as. In the following description, even if the selection can be made by clicking the mouse, the case where the selection is made by the operator pressing the key switch will be described as an example.

As shown in FIG. 4, when the vehicle image shooting process is started, a predetermined initialization process is first performed in step S101, and then a shooting mode setting process is performed in step S103. In the initialization process, for example, the work file, the flag, etc. related to the vehicle image shooting process are cleared, and the setting data, etc. are expanded on the main memory. Further, in the shooting mode setting process, the default values of the setting data related to the zoom, white balance, exposure, etc., which are individually set for each camera 30a and the like, are transmitted to the camera interface 14. As a result, the zoom, white balance, and the like of the camera 30a and the like are set to default values.

In the present embodiment, as in the display screen 70 shown in FIG. 6, for example, the shooting mode is displayed as a camera mode button 96 such as “sunny”, “cloudy”, “evening”, and “night”. For example, "sunny" in the currently set shooting mode is displayed in yellow (light gray in FIG. 7), and other "cloudy" and the like are displayed in gray. In this shooting mode, the operator presses the "camera mode switch" 79 of the F9 key to set the shooting mode to "sunny"-> "cloudy"-> "evening"-> "night"-> "sunny"-> ... Can be switched to cyclic.

When the shooting mode setting process in step S103 is completed, the process waits until there is an instruction to start shooting (S105). That is, as shown in FIG. 6, it waits until the "batch shooting" 75 of the F5 key of the function displayed on the display screen 70 of the computer 11 or the "four-point shooting" 76 of the F6 key is pressed (S105; No). Since the display example of FIG. 6 exemplifies the screen display in a state where the processing has progressed to some extent, the display is not displayed at the time of processing in step S103 (for example, a four-point still image 94 and a multi-still image 95). Note that) is already displayed.

For example, the operator of the computer 11 waits for the start of shooting until the vehicle 8 stops at the predetermined position 7, or waits for the start of shooting until the worker finishes shooting the interior image of the vehicle 8. Further, when the shooting of a moving image is started by the cameras 30a or the like of the camera walls 20a to 20d, the figure of the worker may be reflected in the moving image. Therefore, when the worker completes predetermined work such as moving the vehicle 8 and photographing the interior, the worker hides in the evacuation space 4 (see FIG. 1) provided on the back side of the camera walls 20a to 20d. It is possible to prevent such reflections. The operator presses "Batch shooting" 75 or "Four-point shooting" 76 after confirming by visual inspection or intercommunication that the predetermined work or evacuation by the worker has been completed.

When "Batch shooting" 75 or "Four-point shooting" 76 is pressed (S105; Yes), a control command for starting shooting with an address such as each camera 30a is transmitted from the computer 11 to the camera interface 14. .. As a result, the cameras 30a and the like each start shooting a moving image of the vehicle 8.

The moving image data acquisition process is performed by the subsequent step S107. Here, the case where "Batch shooting" 75 is pressed will be described as an example. When "4 point shooting" 76 is pressed, the video hub 12 is used from the cameras 30b, 30e, 30h, 30k arranged in the center among the three cameras 30a and the like provided in the camera walls 20a to 20d. Only the moving image data sent via the image processing is targeted. This point is different from the case of "batch shooting" in which moving image data of all cameras 30a and the like are targeted for image processing.

In the process of step S107, for example, a control command capable of instructing the video hub 12 to output the video data of the cameras 30a, 30b, 30c among the video data sent from the cameras 30a and the like is given to the video hub 12. Send. As a result, the moving image data of the cameras 30a, 30b, 30c is sent from the video hub 12 via the A channel. As a result, for example, the moving image sent from the cameras 30a, 30b, and 30c is projected on the display of the computer 11 by dividing the screen as shown in the four-point still image 94 shown in FIG. The moving image sent from one specific camera 30 (for example, the camera 30c) may be greatly projected within the broken line indicated by the reference numeral 94.

Therefore, in the next step S109, the image data input from the A channel is captured, the still image data is cut out (captured) from the moving image data, and for example, it is temporarily stored in the storage device. Store. The still image data to be cut out from the moving image data may be one or two or more. When two or more still image data are cut out, the trimming process (S117) described later is also performed on the two or more still image data.

By performing such video data acquisition processing (S107) and capture processing (S109) from channel A to channel D (S111; No), it is possible to cut out still image data for each video data output from the camera 30a or the like. It will be possible. When it is determined in step S111 that the capture process has been performed on all the data (S111; Yes), the shooting completion display process is performed in the subsequent step S113.

The shooting completion display process (S113) is performed by changing the display color of the shooting completion indicator 99a on the display screen 70 from gray to red. That is, in the shooting completion indicator 99a, the background color of the characters "shooting completed" is displayed in gray before the shooting is completed, and the background color is displayed in red after the shooting is completed. The adjustment completion indicator 99b displayed to the right of the shooting completion indicator 99a displays the background color of the characters "adjustment completed" in gray before the adjustment is completed, and the same background color is displayed in red after the adjustment is completed. Will be done.

If a sub-display is provided in the evacuation space 4 behind the camera walls 20a to 20d where the worker can stand by, the shooting completion indicator 99a and the adjustment completion indicator 99b are also displayed on the display screen 70. Is displayed, so the worker can also check it on the screen.

When the still image is cut out (captured) in this way, the trimming frame setting process is then performed in step S115. The trimming in the vehicle image capturing process means to cut out an unnecessary image portion around the image of the vehicle 8 and the boundary line set for that purpose is a trimming frame. The image processing that leaves the inside of the trimming frame and deletes the outside of the frame is the trimming process performed in the next step S117. Trimming is also called "clipping".

The flow of the trimming frame setting process in step S115 is shown in FIG. 5, and an explanatory diagram showing a conceptual image of this process is shown in FIG. Therefore, from here on, it will be described with reference to FIGS. 5 and 7. In addition, FIG. 8 showing the image before trimming and FIG. 9 showing the image after trimming are also referred to.

As shown in FIG. 5, in the trimming frame setting process, first, a predetermined initialization process is performed in step S201. In this process, for example, a default value of a narrowing range described later is set. In the following step S203, the reference image acquisition process is performed. For this reference image, an image taken by a camera capable of taking the widest distance between the wheel (or tire) of the front wheel 9a of the vehicle 8 and the wheel (or tire) of the rear wheel 9b is selected from the cameras 30a and the like. (See Fig. 7 (A)). That is, since the front and rear wheel spacing can be widest in the image taken from the side of the vehicle 8, if a camera is provided near the center pillar of the vehicle 8, the image taken by the camera is used as a reference. Select for the image.

However, in the case of the present embodiment, the camera is not provided near the center pillar of the vehicle 8. Therefore, the images 50c, 50d, 50i, 50j taken by the cameras 30c, 30d, 30i, and 30j, which are closest to the center of the vehicle 8 in the front-rear direction, can correspond to the reference image (FIG. 8 (C), FIG. 8 (D), Fig. 8 (I), Fig. 8 (J)). That is, these cameras 30c, 30d, 30i, and 30j can take the largest image in the front-rear direction of the vehicle 8 as compared with other cameras 30a, 30b, 30e, 30f, 30g, 30h, 30k, and 30m. Unless a part of the vehicle 8 is cut (missing) and photographed by these cameras 30c, 30d, 30i, 30j, a part of the vehicle 8 is cut (missing) in other cameras as well. This is because it cannot occur.

As the target of the image recognition processing, an image that is less affected by external light is more desirable. Therefore, in the image, the image of the entrance 5 in which the range of reflection is narrower than that of the exit 6 in which the light enters from the outside is incident, that is, the image 50c or the image 50j is used as the reference image. Select to. In the present embodiment, the image 50j taken by the camera 30j is used as the reference image. In this specification, a camera that has taken a reference image may be referred to as a reference camera.

Further, in the next step S205, the wheel detection process is performed. This process recognizes a circle or an ellipse at two points from the image data of the reference image (an image including a vehicle image in which the side surface of the vehicle appears) and detects the images of the wheels of the front wheels 9a and the rear wheels 9b. The wheels of the front wheels 9a and the rear wheels 9b appear as a circular (circular) image in the image taken from the side of the vehicle 8 and as an elliptical (elliptical) image in the image taken from the diagonal side of the vehicle 8. .. Therefore, in the image 50j (reference image) taken by the camera 30j, the wheels of the front wheels 9a and the rear wheels 9b appear as elliptical images. Therefore, in the image recognition of the wheel detection process in step S205, for example, they are registered in advance. Template pattern matching is performed based on multiple types of elliptical templates corresponding to the wheels.

Further, in the present embodiment, the default value of the narrowing range is set by the initialization process of step S201. Therefore, the wheel is not detected by image recognition for the entire range of the reference image, but the wheel is detected within the narrowing range (narrowing frame α). That is, in the present embodiment, for example, as shown in FIG. 7B, the narrowing down that surrounds a range smaller than the entire image range α of the image 50j (within the gray colored outer peripheral range shown in FIG. 7B). Wheel detection is performed in the frame β. The narrowing frame β is based on the positional relationship between the angle of view (shooting range) of the camera 30j that captured the image 50j and the predetermined position 7, and when the vehicle 8 stops at the predetermined position 7, the front wheel 9a of the vehicle 8 is used. And the rear wheel 9b is generally set to the position and range.

If the front wheels 9a and the rear wheels 9b do not fit within the range of the narrowing frame β depending on the size of the vehicle 8, another narrowing frame having a size larger than that of the narrowing frame β is called from the storage device, and step S205 is performed. Used for wheel detection processing. When the front wheels 9a and the rear wheels 9b do not fit within the narrowing frame β, when only one ellipse (or circle) is detected in the narrowing frame β, or when no ellipse (or circle) is detected. Is.

When the images of the wheels of the front wheels 9a and the rear wheels 9b are detected in step S205, the wheelbase calculation process is performed in step S207. The wheelbase is the distance between the axle of the front wheel 9a and the axle of the rear wheel 9b of the vehicle 8. Therefore, in this wheelbase calculation process, the centers 53 and 54 of the ellipses (or circles) 51 and 52 of the front wheels 9a and the rear wheels 9b image-recognized by the wheel detection process in step S205 are obtained, and then the centers of both are obtained. The distance between them, that is, the wheelbase 57 is calculated. The wheelbase (distance between centers) 57 is represented by, for example, the number of pixels.

The wheelbase 57 calculated in step S207 is used for the trimming frame determination process in the next step S209. Generally, the total length of the vehicle 8 is often influenced by the wheelbase 57, so that the total length of the vehicle 8 is roughly determined by the size of the wheelbase 57. That is, it is possible to determine the size (size (total length, total width, total height)) of the vehicle 8 corresponding to each length of the wheelbase 57. Therefore, in the present embodiment, a plurality of types of trimming frames γ that can surround the entire vehicle of each size (total length, total width, total height) are associated with each length of the wheelbase 57. For example, the frame size buttons 93 (“frame 1” to “frame 9”) displayed on the display screen 70 of FIG. 6 are for the operator to manually select such a plurality of types of trimming frames γ. be.

When the trimming frame γ that can surround the entire vehicle 8 is determined (set) by the trimming frame determination process in step S209, the present trimming frame setting process is terminated and the process returns to the vehicle image capturing process shown in FIG.

In the trimming process in step S117, after cutting out from the moving image data of the camera 30a or the like and reading the still image data temporarily stored in the storage device or the like, the trimming frame γ set in step S115 is used in the same frame γ. It is a process to cut out the image data of the above and generate another image data. This process is performed until the end of all the still image data cut out in step S109 (S119; No). As a result, as will be described later, by connecting the still image data after the trimming process in step S117 so as to be arranged in the order of the peripheral direction of the vehicle 8, it is possible to see the surroundings of the vehicle body around the entire circumference of the vehicle 8 (around view). It can be seen that the vehicle body image data equivalent to (360 degree image) can be generated.

Note that FIG. 8 shows an example of image data before the trimming process (S117). Further, FIG. 9 shows an example of image data after the trimming process (S117). From these figures, it can be understood that the image and the like after the trimming of FIG. 9 are cut out from the image 50a and the like before the trimming of FIG. It can also be seen that the images 60a and the like in FIG. 9 are the still images displayed in the multi-still image 95 of the display screen 70 shown in FIG.

Further, the sizes Xb and Yb of the trimming frame γ are rectangles suitable for the display size of the display screen 70 for displaying the trimmed image 60a and the like (for example, 1280 pixels × 960 pixels with an aspect ratio of 4: 3 corresponding to QVGA). It is set to the size. The sizes Xa and Ya of the image 50a and the like before trimming are set to, for example, 1920 pixels × 1080 pixels having an aspect ratio of 16: 9 corresponding to full HD.

When it is determined in step S119 that the trimming process has been performed on all the data (S119; Yes), the vehicle body image display process is performed in the subsequent step S121. In this process, the still image after trimming in step S119 is displayed on the four-point still image 94 and the multi-still image 95 of the display screen 70. That is, the content shown in FIG. 6 is displayed on the display screen 70 of the computer 11. That is, in the four-point still image 94, the upper right image is taken by the camera 30b, the lower right image is taken by the camera 30e, the lower left image is taken by the camera 30h, and the upper left image is taken by the camera 30k.

Further, in the multi-still image 95, for convenience of explanation, when 12 images are divided into four vertically and horizontally, the left image is the camera 30a, the center image is the camera 30b, and the right image is the camera 30c among the three upper right images. Of the three lower right images, the right image was taken by the camera 30d, the center image was taken by the camera 30e, and the left image was taken by the camera 30f. Of the three lower left images, the right image is taken by the camera 30 g, the center image is taken by the camera 30h, and the left image is taken by the camera 30i. Of the upper left three images, the left image is the camera 30j and the center image is the camera 30k. , The right image is taken by the camera 30m respectively.

By connecting the still images displayed in the multi-still image 95 so as to be arranged in the order of the surrounding direction of the vehicle 8, the vehicle body image of the entire circumference of the vehicle 8 can be viewed as an around view (360 degree image). It is possible to show a pseudo-rotation as if it were rotating. Further, as shown in FIG. 9, in the vehicle image shown in each image 60a or the like, the roof position Yi of the vehicle 8 is uniformly aligned in the height direction, so that the roof position is in the height direction. Compared to the case where they are disjointed, the position of the roof due to such pseudo rotation is less likely to be uneven. Therefore, the pseudo rotation of the vehicle image can be made to look beautiful.

That is, the upper right left image of the camera 30a (FIG. 9 (A); 60a) → the upper right center image of the camera 30b (FIG. 9 (B); 60b) → the upper right image of the camera 30c (FIG. 9 (C); 60c) → Lower right right image of camera 30d (FIG. 9 (D); 60d) → Lower right center image of camera 30e (FIG. 9 (E); 60e) → Lower right left image of camera 30f (FIG. 9 (F); 60f) → Lower left right image of camera 30g (Fig. 9 (G); 60g) → Lower left center image of camera 30h (Fig. 9 (H); 60h) → Lower left left image of camera 30i (Fig. 9 (I); 60i) → Camera The upper left left image of 30j (FIG. 9 (J); 60j) → the upper left center image of the camera 30k (FIG. 9 (K); 60k) → the upper left right image of the camera 30m (FIG. 9 (L); 60m). Each still image data is displayed sequentially in. This makes it possible to obtain vehicle body image data equivalent to an around view (360-degree image).

When "four-point shooting" 76 is pressed in step S105, only the four-point still image 94 is displayed on the display screen 70 and the multi-still image 95 is not displayed in step S121. Therefore, when "4 point shooting" 76 is pressed, the vehicle body image data equivalent to the around view (360 degree image) cannot be obtained.

In the following step S123, the adjustment completion display process is performed. In this process, the display color of the adjustment completion indicator 99b on the display screen 70 is changed from gray to red. As described above, the adjustment completion indicator 99b has a gray background color before the adjustment is completed.

When such a series of capture (cutout) and trimming (cutout) is completed, the computer 11 waits for the next input command in step S125. That is, it is determined whether the "save" 83 of the ENTER key, the "screen clear" 71 of the ESC key, or the "end" 82 of the F12 key is pressed.

When it is determined that "Save" 83 is pressed (S125; "Save"), all the image data of the still image after trimming by the image data saving process of the following step S127 is saved in the file server 13. As a result, unique information about the vehicle 8 such as the exhibition number associated with the saved still image data is displayed in the storage history display area 97 of the display screen 70. Then, after erasing the four-point still image 94, the multi-still image 95, etc. currently displayed by the image display erasing process in step S129, the process is shifted to step S105 to prepare for the next shooting start, "batch shooting". Wait for the keys of 75 and "4 point shooting" 76 to be pressed.

On the other hand, when it is determined that the "screen clear" 71 is pressed (S125; "screen clear"), the image data saving process in step S127 is skipped and the image display erasing process in step S129 is performed. That is, after discarding the trimmed still image data without saving (S129), the process shifts to step S103 and the keys of "Batch shooting" 75 and "Four-point shooting" 76 are pressed in preparation for the next shooting start. Wait for it to be pressed.

When "End" 82 is pressed (S125; "End"), for example, a separate window or the like for confirming whether or not the vehicle image shooting process can be completed is displayed on the display screen 70 for final confirmation. When is input, the vehicle image shooting process is finished and the display of the display screen 70 is finished. The pressing of "End" 82 can be accepted even while other processing is being executed.

As a process not described in the flowchart of FIG. 4, for example, when the "interior redisplay" 72 of the F2 key of the function is pressed, the interior image of the vehicle 8 is redisplayed in the interior image display area 92. The interior image data (interior image data) is photographed by a digital camera carried by a worker separately from the cameras 30a and the like provided on the camera walls 20a to 20d, and is stored in the file server 13 via a wireless LAN or the like. It is also displayed in the interior image display area 92 of the display screen 70 at almost the same time or at another timing. The interior image data is associated with data unique to the vehicle 8 such as an exhibition number. Therefore, when the "interior redisplay" 72 is pressed, the computer 11 acquires the interior image data of the vehicle 8 from the file server 13 based on the unique data of the vehicle 8 and displays it again in the interior image display area 92. ..

The interior image deletion area 91 provided to the left of the interior image display area 92 uses a mouse or the like to capture an interior image that is unnecessary when deleting the interior image displayed in the interior image display area 92. It is used to exclude the interior image from the interior image display area 92 by dragging and dropping it onto the image deletion area 91.

Further, the "retrimming" 73 of the F3 key of the function is, for example, when the F4 key 74 of the function is pressed and the setting of the trimming frame is switched from "automatic" to "manual", when the trimming frame is changed. Pressed by the operator. When the "retrimming" 73 is pressed, the computer 11 reads the temporarily stored still image data from the storage device or the like, and then uses the newly set trimming frame to display the image data in the frame. Performs cropping image processing.

The F4 key 74 of the function enables the manual setting and the automatic setting of the trimming frame to be selected by toggle (alternate each time the button is pressed) by pressing the F4 key 74. That is, when the manual setting is selected, "manual" is displayed in the part of the F4 key 74, and when the automatic setting is selected, "automatic" is displayed in the part of the F4 key 74. .. The vehicle image shooting process described above is premised on the automatic setting of the trimming frame.

Further, when the "multi-shooting" 77 of the F7 key of the function is pressed, the computer 11 displays the 12 still images trimmed by the trimming process (S117) in a separate window or the like at four points of still images 94. Display as large as. In this case, it is difficult to display all 12 still images at the same time on the display screen 70. Therefore, for example, as described above, a still image that is enlarged by operating the scroll wheel of the mouse or a specific key of the keyboard is displayed as the upper right left image of the camera 30a → the upper right center image of the camera 30b → the upper right right image of the camera 30c → the camera. Lower right right image of 30d → Lower right center image of camera 30e → Lower right left image of camera 30f → Lower left right image of camera 30g → Lower left center image of camera 30h → Lower left left image of camera 30i → Upper left left image of camera 30j → Perform display processing to change the display image so that it appears in the order of the upper left center image of the camera 30k → the upper left right image of the camera 30m.

Further, when the "exhibition number acquisition" 78 of the F8 key of the function is pressed, the computer 11 acquires the exhibition number of the vehicle 8 from the file server 13 based on the data unique to the vehicle 8, and the exhibition number. Display it in the display area 98. Further, after the operator clicks the exhibition number display area 98 with a mouse or the like, the exhibition number of the vehicle 8 can be directly input with the keyboard. The exhibition number is associated with the still image data when the still image data is stored in the file server 13.

When the trimming frame is set to "manual", the button portion of the size currently set as the frame size button 93 is displayed in yellow (light gray in FIG. 6), and the other button portions are gray. It is displayed in. This yellow display goes up or down each time the "frame size +" 80 of the F10 key or the "frame size-" 81 of the F11 key is pressed. The trimming frame can also be selected by clicking the frame size buttons 93 of "frame 1" to "frame 9" with a pointing device such as a mouse.

Next, the types of lenses such as the cameras 30a provided on the camera walls 20a to 20d will be described. In the present embodiment, not all lens units such as the camera 30a are selected. That is, the camera 30a or the like may have a standard zoom lens or a wide-angle zoom lens.

For example, a wide-angle zoom lens is selected for the cameras 30c, 30d, 30i, and 30j, which are arranged at the positions closest to the vehicle 8 in the peripheral direction of the vehicle 8 among the three cameras of the camera walls 20a to 20d. To. A wide-angle zoom lens is selected for the cameras 30b, 30e, 30h, and 30k, which are arranged in the center of the vehicle 8 in the peripheral direction among the three camera walls 20a to 20d. Then, a standard zoom lens is selected for the remaining cameras 30a, 30f, 30g, and 30m.

Of the three cameras 30 provided on the camera walls 20a to 20d, the cameras 30a, 30f, 30g, and 30m arranged at the farthest end from the vehicle 8 have a standard zoom lens instead of a wide-angle zoom lens. Standard zoom lenses generally tend to be cheaper than wide-angle zoom lenses. Therefore, it is possible to suppress an increase in the cost of the photographing system 10 as compared with the case where a wide-angle zoom lens is selected for all the cameras 30a and the like.

As described above, in the photographing system 10 of the present embodiment to which the vehicle image generation device and the vehicle image generation program of the present invention are applied, the wheel detection process (S205) is performed by two ellipses 51 from the image data including the vehicle image 40. , 52 (or circle) is recognized, and the wheelbase calculation process (S207) measures the distance between the centers 53, 54 of the two ellipses 51, 52 (or circle) recognized by the wheel detection process (S205). .. The trimming frame determination process (S209) determines a trimming frame γ that can determine a predetermined range in which the vehicle image 40 exists based on the distance between the centers 53 and 54, and the trimming process (S117) determines the trimming frame γ. It is used to crop an image within a predetermined range in which the vehicle image 40 exists to generate another image data. As a result, the trimming process (S117) cuts out the image 60j within a predetermined range using such a trimming frame γ, so that the vehicle image 40 has the size of the trimming frame γ without detecting the contour of the vehicle image 40. Can be cropped to generate different image data.

Further, in the photographing system 10 of the present embodiment, the wheel detection process (S205) recognizes two ellipses 51, 52 (or circles) in the narrowing frame β smaller than the entire image range α such as the image 50a. As a result, the target range of the wheel detection process (S205) by the computer 11 is narrower than in the case of recognizing such two ellipses 51, 52 (or circles) in the entire image range α such as the image 50a. The time required to recognize the two ellipses 51, 52 (or circles) is shortened. Therefore, the wheel detection process (S205) can be speeded up.

<< Other reference examples of trimming frame setting process >>
If the unique information (exhibition number, vehicle type, model, displacement, mileage, vehicle name, etc.) regarding the vehicle 8 described in the exhibition inspection slip is stored (stored) in the file server 13, the display screen is displayed. The size of the vehicle 8 can be called from the file server 13 from the exhibition number of the vehicle 8 displayed in the exhibition number display area 98 of the 70. The unique information about the vehicle 8 can correspond to the "vehicle information" described in the claims. Therefore, in such a case, in the trimming frame setting process (S115), the process shown in FIG. 10 can be configured. FIG. 10 shows a flowchart showing the flow of another reference example of the subroutine “trimming frame setting process” shown in FIG.

As shown in FIG. 10, in the trimming frame setting process, first, a predetermined initialization process is performed in step S501. In the following step S503, the exhibition number acquisition process is performed. As described above, when the "exhibition number acquisition" 78 of the F8 key of the function of the display screen 70 is pressed, the computer 11 determines the exhibition number from the file server 13 to the vehicle 8 based on the data unique to the vehicle 8. The exhibition number is acquired and displayed in the exhibition number display area 98. Further, after the operator clicks the exhibition number display area 98 with a mouse or the like, the exhibition number of the vehicle 8 is directly input with the keyboard.

Since the exhibition number is associated with unique information about the vehicle 8 (exhibition number, vehicle type, model, displacement, mileage, vehicle name, etc.), the vehicle name or the vehicle name or the vehicle name is obtained by the vehicle size acquisition process in the next step S505. The size of the vehicle 8 is acquired based on the model. The size of the vehicle 8 is acquired, for example, by reading from the vehicle master stored in the file server 13. As a result, data such as the total length, the total width, and the total height of the vehicle 8 can be obtained, so the trimming frame is determined in the subsequent step S507. It suffices if at least the total length can be obtained.

The trimming frame determination process in step S507 is substantially the same as step S209 in FIG. 5 described above. Of the sizes of the vehicle 8 (total length, total width, total height), the trimming frame γ associated with each total length is determined. The trimming frame γ can surround the entire vehicle of the vehicle 8 in the image 50a or the like. When the trimming frame γ that can surround the entire vehicle 8 is determined (set) by the trimming frame determination process in step S507, the present trimming frame setting process is terminated and the process returns to the vehicle image capturing process shown in FIG.

As described above, in another reference example of the trimming frame setting process in the photographing system 10 of the present embodiment, in the vehicle size acquisition process (S505), the unique information about the vehicle 8 that can specify the total length of the vehicle 8 is acquired. Then, in the trimming frame determination process (S507), a trimming frame γ that can determine a predetermined range in which the vehicle image 40 exists is determined based on the total length included in the unique information of the vehicle 8, and in the trimming process (S117), the trimming frame γ is determined. Using the trimming frame γ, the image 60j within a predetermined range in which the vehicle image 40 exists is cut out to generate another image data.

As a result, based on the total length included in the unique information about the vehicle 8 acquired by the vehicle size acquisition process (S505), the trimming frame determination process (S507) can determine a predetermined range in which the vehicle image 40 exists. Can be determined based on the unique information about the vehicle 8. Therefore, the trimming process (S117) cuts out the image 60j within a predetermined range using such a trimming frame γ, so that the vehicle image 40 is cut out to a predetermined size without detecting the contour of the vehicle image 40. Another image data can be generated.

In the above-described embodiment, two ellipses 51, 52 (or circles) are recognized by the wheel detection process (S205) for the image (still image captured from the moving image) 50j taken by the camera 30j as a reference camera. The two ellipses 51, 52 (or circles) may be recognized by the wheel detection process (S205) for the image 50c taken by using the camera 30c as a reference camera. Further, depending on the conditions of the external light incident from the inlet 5 and the exit 6, two ellipses 51, 52 (or circles) are formed by the wheel detection process (S205) for the image 50d taken by the camera 30d as a reference camera. It may be recognized, or two ellipses 51, 52 (or circles) may be recognized by the wheel detection process (S205) for the image 50i taken by using the camera 30i as a reference camera.

Further, in the above-described embodiment, the images 50a and the like before trimming are photographed by the camera 30a and the like so that the heights of the uppermost roof contours of the vehicle 8 are uniformly aligned, and these images 50a are subjected to the trimming process (S117). Etc. are uniformly trimmed at the same trimming position, but in the trimming process, the image 50a or the like before trimming may be trimmed so that the heights of the uppermost roof contours of the vehicle 8 are uniformly aligned. In this case, of the front wheels 9a or the rear wheels 9b image-recognized by the image-recognized wheel detection process (S205), the center 53 (or the center 54) of one of the front wheels 9a (or the rear wheel 9b) located below in the image. ) Is used as a reference to determine the vertical position of the trimming frame.

Although specific examples of the present invention have been described in detail above, these are merely examples and do not limit the scope of claims. The techniques described in the claims include various modifications or modifications of the above-mentioned specific examples. Further, the technical elements described in the present specification or the drawings exhibit technical usefulness alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. Further, the techniques exemplified in the present specification or the drawings achieve a plurality of objectives at the same time, and achieving one of the objectives has technical usefulness in itself. The description in parentheses in the [Explanation of reference numerals] column can clearly indicate the correspondence between the terms used in each of the above-described embodiments and the terms described in the claims.

2 ... Shooting studio 4 ... Evacuation space 5 ... Entrance 6 ... Exit 7 ... Predetermined position 8 ... Vehicle 8a ... Front side 8b ... Rear side (side of vehicle)
8c ... Left side (side of the vehicle)
8d ... Right side 9a ... Front wheel 9b ... Rear wheel 10 ... Shooting system 11 ... Computer (vehicle image generator, image recognition means, distance measuring means, frame determining means, image cropping means)
20, 20a to 20d ... Camera wall 30, 30a to 30k, 30m ... Camera 40 ... Vehicle image 50 ... Image 50c, 50d, 50i, 50j ... Image (image including vehicle image showing the side surface of the vehicle)
51, 52 ... Ellipse 53, 54 ... Center 57 ... Wheelbase (distance between the centers of a circle or ellipse)
60 ... Image 70 ... Display screen S117 ... Trimming process (image cropping means)
S203 ... Reference image acquisition processing S205 ... Wheel detection processing (image recognition means)
S207 ... Wheelbase calculation process (distance measuring means)
S209 ... Trimming frame determination process (frame determination means )
α ... Full range of image β ... Narrowing frame γ ... Trimming frame

Claims (3)

A vehicle image generation device that generates another image data by cutting out an image within a predetermined range in which the vehicle image exists from image data including a vehicle image in which a side surface of the vehicle appears.
An image recognition means that recognizes two circles or ellipses from the image data,
A distance measuring means for measuring the distance between the centers of the two circles or ellipses recognized by the image recognition means, and a distance measuring means.
A frame determining means for determining a predetermined frame that can determine the predetermined range based on the distance between the centers measured by the distance measuring means, and
An image cropping means that cuts out an image within the predetermined range using the predetermined frame determined by the frame determining means to generate the other image data, and an image cropping means.
A vehicle image generator, characterized in that it comprises. The vehicle image generation device according to claim 1, wherein the image recognition means recognizes the two circles or ellipses within a range smaller than the entire image range of the image data. It is a vehicle image generation program that generates another image data by cutting out an image within a predetermined range in which the vehicle image exists from image data including a vehicle image in which a side surface of the vehicle appears.
Computer,
An image recognition means that recognizes two circles or ellipses from the image data,
A distance measuring means for measuring the distance between the centers of the two circles or ellipses recognized by the image recognition means, and a distance measuring means.
A frame determining means for determining a predetermined frame that can determine the predetermined range based on the distance between the centers measured by the distance measuring means, and
An image cropping means that cuts out an image within the predetermined range using the predetermined frame determined by the frame determining means to generate the other image data, and an image cropping means.
A vehicle image generation program characterized by functioning.

Images