JP2014016878A - Adjustment method of imaging device, imaging device, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily set a parameter of a camera properly.SOLUTION: First, an imaging unit 111 picks up an image including a part of a rectangular area on a road. Then, a coordinate specification unit 114 specifies coordinates of each of vertices of the rectangular area on the basis of the image picked up by the imaging unit 111. Then, a parameter specification unit 115 specifies a parameter on the basis of a shape of a closed section connecting the coordinates specified by the coordinate specification unit 114. Then, referring to the parameter specified by the parameter specification unit 115, in order to cause the parameter to be closer to a target parameter, the parameter is adjusted by adjusting a zoom mechanism 121, a horizontal rotation mechanism 133, a vertical rotation mechanism 134, and a vertical expansion/contraction mechanism 135.

Description

  The present invention relates to an imaging device adjustment method, an imaging device, and a program.

  A conventional ERP (Electronic Road Pricing) system includes an imaging device in a first gantry device provided in a passage, and a second gantry device provided in front of the passage in the vehicle traveling direction from the first gantry device. A vehicle detector is provided. In such an ERP system, when the vehicle detector detects the vehicle, the imaging device images the vehicle, and a CCS (Central Computer System) performs license plate recognition processing on the captured image. (For example, refer to Patent Document 1).

  In recent years, development of an ERP system (hereinafter referred to as a next-generation ERP system) using GPS (Global Positioning System) or third-generation mobile communication is planned. In the next-generation ERP system, since the vehicle recognition process is performed using GPS or mobile communication, the antenna and the gantry device installed in the passage in the conventional ERP system become unnecessary.

  In the next-generation ERP system, since a gantry device is not required, a configuration in which the vehicle detector provided in the second gantry device in the conventional ERP system is not used has been studied. Therefore, a vehicle information recognition device for recognizing a vehicle number (vehicle information) written on a license plate of the vehicle for the control of an unauthorized vehicle is installed on the road side. In this case, it is conceivable that the vehicle information recognition device is fixedly installed on roadside equipment such as a cantilever or a pole, but for ease of maintenance and use, the vehicle information recognition device can be installed on a tripod to make it portable. There is a desire to do so (see, for example, Patent Document 2).

JP 2002-260165 A Japanese Patent Laid-Open No. 08-055296

  However, even if the portable vehicle information recognition device as described above is installed, in order for the vehicle shown in the captured image to have an appropriate angle and size for recognition of vehicle information, However, it is necessary to finely adjust parameters such as the imaging direction within a smooth plane and the imaging depression angle and angle of view within a vertical plane. However, it is not easy to properly adjust these parameters in the field.

  The present invention has been made in view of the above-described problems, and provides an imaging apparatus adjustment method, an imaging apparatus, and a program that can appropriately and easily set parameters of the imaging apparatus.

The present invention has been made to solve the above-described problem, and is an imaging device adjustment method in which an imaging unit of the imaging device captures an image including a part of a rectangular region on a road surface. A coordinate specifying step in which the coordinate specifying unit of the imaging device specifies the coordinates of each vertex of the rectangular area based on the image captured by the imaging unit, and the parameter specifying unit of the imaging device includes the coordinate specifying unit The parameter specifying step for specifying the parameter based on the shape of the closed section connecting the specified coordinates and the parameter specified by the parameter specifying unit, and adjusting the imaging device so that the parameter approaches the target parameter And an adjusting step for adjusting the portion.
The “parameter” (camera parameter) is various setting items on the camera side for adjusting the captured image. The imaging direction on the horizontal plane (left and right imaging directions), the imaging direction on the vertical plane (upper and lower imaging) Direction), angle of view, camera installation position, focal length, focus state, aperture state, camera shutter speed, camera sensitivity, optical filter characteristics, and the like.

  Moreover, in this invention, it is preferable that the said rectangular area | region uses the line of the both ends of a lane, and the line orthogonal to the said line as a side.

  Further, in the present invention, before the imaging step, there is a widening step of adjusting the angle of view by the adjustment unit of the imaging device so that the image captured by the imaging unit includes all of the rectangular area, In the adjustment step, it is preferable to perform an adjustment to make the angle of view close to the target parameter after performing an adjustment to make the direction angle and the depression angle close to the target parameter among the parameters.

  Further, the present invention is based on an imaging unit that captures an image including a part of a rectangular area on a road surface, an adjustment unit that adjusts a parameter that determines an imaging range of the imaging unit, and an image captured by the imaging unit. A coordinate specifying unit that specifies the coordinates of each vertex of the rectangular region, and a parameter specifying unit that specifies the parameter based on the shape of a closed section connecting the coordinates specified by the coordinate specifying unit. An imaging device.

  In the present invention, vehicle information is recognized from an image captured by the imaging unit using setting information designed to provide the best recognition accuracy when the imaging range determined by the target parameter is imaged. It is preferable to provide a recognition unit.

  Moreover, in this invention, it is preferable that the said rectangular area | region uses the line of the both ends of a lane, and the line orthogonal to the said line as a side.

Moreover, in this invention, it is preferable to provide the presentation part which compares the parameter which the said parameter specific | specification part specified with the predetermined target parameter, and shows the comparison result.
Examples of the “presentation unit” include a display that displays characters and images indicating the comparison result, a speaker that outputs sound indicating the comparison result, and the like.

  In addition, the present invention provides an imaging device that captures an image including a part of a rectangular area on a road surface, and coordinate specification that specifies the coordinates of each vertex of the rectangular area based on the image captured by the imaging section And a program for functioning as a parameter specifying unit for specifying the parameter based on the shape of a closed section connecting the coordinates specified by the coordinate specifying unit.

  According to the present invention, since the parameter specifying unit of the image pickup apparatus specifies the current parameter, the user adjusts the adjustment unit so that the current parameter and the target parameter substantially match, whereby the image pickup apparatus The parameters can be set appropriately and easily.

1 is an external view of a vehicle information recognition device according to an embodiment of the present invention. It is a figure which shows the designation | designated method of the rectangular area used by one Embodiment of this invention. It is a flowchart which shows the adjustment method of the parameter of the vehicle information recognition apparatus by one Embodiment of this invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is an external view of a vehicle information recognition apparatus 100 according to an embodiment of the present invention.
As shown in FIG. 1, the vehicle information recognition device 100 (imaging device) has a camera 110 that captures a moving image, a zoom lens 120 that adjusts the angle of view (zoom amount) of the camera 110, and a camera 110 as main components. A tripod 130 is provided.

  The zoom lens 120 and the tripod 130 are provided with an adjustment unit for adjusting parameters (such as depression angle, direction angle, and field angle) that determine the imaging range of the camera 110. Specifically, the zoom lens 120 is provided with a zoom mechanism 121 that is an adjustment unit for adjusting the angle of view, and the angle of view can be adjusted steplessly. The tripod 130 has three legs 131 standing on the ground and a camera support 132 to which the camera 110 is mounted. The legs 131 and the camera support 132 adjust the direction angle. They are connected by a horizontal rotation mechanism 133 that is an adjustment unit, a vertical rotation mechanism 134 that is an adjustment unit that adjusts the depression angle, and a vertical telescopic mechanism 135 that is an adjustment unit that adjusts the height. The horizontal rotation mechanism 133 rotates the camera 110 mounted on the camera support unit 132 in a substantially horizontal plane, and the vertical rotation mechanism 134 rotates the camera 110 mounted on the camera support unit 132 in a substantially vertical plane. Rotate in.

  The camera 110 is equipped with a computer for controlling imaging by the camera 110. The computer includes an imaging unit 111, a recognition unit 112, an I / F unit 113, a coordinate specification unit 114, a parameter specification unit 115, a target parameter storage unit 116, and a presentation unit 117 by executing a predetermined program.

The imaging unit 111 captures a moving image with the camera 110.
The recognition unit 112 recognizes vehicle information using an algorithm that compares a comparison template (setting information), which is a reference character image, with a character image included in the moving image with respect to the moving image captured by the imaging unit 111. Do.
The I / F unit 113 outputs information to the user via an output device such as a display or a speaker, and receives information input from the user via the input device. The I / F unit 113 is realized by, for example, a touch panel.
The coordinate specifying unit 114 receives input of coordinates of four vertices of a rectangular area indicating a part of a lane in a moving image captured by the imaging unit 111 from the user via the I / F. Note that the shape of the rectangular area on the image captured by the imaging unit 111 is not necessarily rectangular.

FIG. 2 is a diagram showing a method for specifying a rectangular area used in an embodiment of the present invention.
In this embodiment, as a rectangular area, as shown in FIG. 2, a rectangle whose sides are lines A and B at both ends of a lane through which a vehicle as a vehicle information recognition target passes and lines C and D orthogonal to the lane. And the area | region which makes the length of a depth direction (A, B) the length of one line segment of the lane boundary printed on the road surface with the broken line is used. That is, the rectangular area is about 12 meters (the length of one line segment of the lane boundary line) × about 3.6 meters (the length of the width of the lane).

The parameter specifying unit 115 specifies the current parameter value based on the vertex from which the coordinate specifying unit 114 has received an input.
The target parameter storage unit 116 stores a target parameter that is a parameter for imaging an imaging range suitable for the vehicle information recognition algorithm executed by the recognition unit 112.
The presenting unit 117 presents the difference value (comparison result) between the parameter identified by the parameter identifying unit 115 and the target parameter stored in the target parameter storage unit 116 to the user via the I / F unit 113.

Here, the target parameter and the comparison template will be described. In advance, the user adjusts the imaging conditions of the camera so that the characters on the license plate are easy to read, and sets various camera parameters under the conditions as “target parameters”. In addition, a character image cut out from the license plate image picked up with the target parameter becomes a “comparison template” (standard character image). These are recorded in the storage unit 116.
For example, the closer the license plate size (range), character size (height, width, line thickness), and character spacing on the plate is closer to that of the comparison template, Since the degree of coincidence between the comparison template and the photographed character image is high, the recognition accuracy of the vehicle information (information described on the license plate) is high.
That is, the recognition unit 112 can recognize the vehicle information with the highest accuracy when the imaging unit 111 images the imaging range determined by the target parameter stored in the target parameter storage unit 116.

Next, a parameter adjustment method for the vehicle information recognition apparatus 100 according to an embodiment of the present invention will be described.
FIG. 3 is a flowchart illustrating a parameter adjustment method of the vehicle information recognition apparatus 100 according to an embodiment of the present invention.

  First, the user adjusts the direction angle and the depression angle to be the target parameters by the horizontal rotation mechanism 133 and the vertical rotation mechanism 134 of the tripod 130 and installs them on the road side (step S1). At this time, the direction angle and the depression angle do not need to exactly match the target parameter. Next, the user monitors the moving image picked up by the image pickup unit 111, so that the entire predetermined rectangular area can be accommodated in the moving image by the vertical extension mechanism 135 of the tripod 130 and the zoom mechanism 121 of the zoom lens 120. Then, the height and the angle of view are adjusted to each other (step S2).

  Next, the user refers to the image displayed by the I / F unit 113 of the vehicle information recognition apparatus 100, and inputs the two-dimensional coordinates of the four vertices of the rectangular area on the image (step S3). For example, the user may input coordinates by touching a corresponding point on the touch panel, or may input coordinates by pressing a direction key.

  When the user inputs the coordinates of all the vertices of the rectangular area, the coordinate specifying unit 114 of the vehicle information recognition apparatus 100 receives the input of the coordinates of the vertices via the I / F unit 113. At this time, the user first inputs the coordinates of both ends of the lane boundary line painted on the road with a broken line among the lanes whose vehicle numbers are to be recognized, and then the corresponding point of the other vehicle boundary line. By inputting the coordinates, the rectangular area can be easily specified.

  Next, the parameter specifying unit 115 specifies the parameters of the camera 110 based on the center point of the rectangular area based on the four coordinates specified by the coordinate specifying unit 114 (step S4). That is, the parameter specifying unit 115 calculates a parameter that forms a closed section connecting the four coordinates specified by the coordinate specifying unit 114 when an image of a rectangular area of 12 meters × 3.6 meters is captured.

Specifically, the parameter specifying unit 115 specifies the relative position (x coordinate, y coordinate, z coordinate) of each vertex of the rectangular area when the center point of the rectangular area is used as a reference (origin). Next, the parameter specifying unit 115 includes a camera matrix P _ε which is a 3 × 4 matrix indicating parameters, a specified relative position vector X _ε , and coordinates on the screen specified by the coordinate specifying unit 114 (x coordinate, y coordinate). The value of the parameter is specified by solving the equation shown in Expression (1) using the vector u of). Here, the camera matrix P _ε is the angle of view (focal length), aspect ratio, non-orthogonal distortion, image center, parallel movement amount in the X, Y, and Z directions indicating the position of the camera 110, and the attitude of the camera 110. Is a matrix including a depression angle and a direction angle. Of the camera matrix P _epsilon, depression and direction angle, X-direction, parallel movement amount in the Y direction and the Z direction, and angle of view, an unknown (certain value is required) parameters.
The x coordinate, the y coordinate, and the z coordinate indicate coordinates based on the coordinate axis of the virtual space that is a three-dimensional Euclidean space, and the X direction, the Y direction, and the Z direction indicate the coordinate axes of the real space, respectively. Indicates the reference movement direction.

_{P ε X ε × u = 0} ··· (1)
Here, “×” is an outer product operator.

Expression (1) shows a state in which a vector X _ε that is a reference point in the real space is projected on the virtual space by the camera matrix P _ε and a vector u that is a corresponding in the virtual space match. It is a formula.

  When the parameter identification unit 115 identifies the parameter of the camera 110 in step S4, the presentation unit 117 calculates a difference value between the parameter identified by the parameter identification unit 115 and the target parameter stored in the target parameter storage unit 116, and I / The difference value is presented to the user via the F unit 113 (step S5). Next, the user refers to the difference value between the direction angle and the depression angle displayed on the I / F unit 113, and determines whether or not the value is within an error range (step S6). For example, when the difference value between the directional angle and the depression angle is within 2 degrees, it can be determined that it is within the error range.

  When the user determines that at least one of the difference value of the direction angle and the difference value of the depression angle is not within the error range (step S6: NO), the horizontal rotation mechanism 133 and the vertical rotation mechanism of the tripod 130 are displayed. By 134, the direction angle and the depression angle are finely adjusted so as to approach the target parameter (step S7). And it returns to step S3 again and inputs the coordinate of a rectangular area.

  On the other hand, when the user determines that both the difference values of the direction angle and the depression angle are within the error range (step S6: YES), the user refers to the difference value of the angle of view displayed on the I / F unit 113, The angle of view is adjusted so as to approach the target parameter by the zoom mechanism 121 of the zoom lens 120 (step S8). When the adjustment of the angle of view is completed, the user starts execution of vehicle information recognition processing by the recognition unit 112 (step S9). Specifically, the recognition unit 112 extracts a frame including a vehicle by performing existing image processing such as pattern matching on each frame of the moving image captured by the imaging unit 111, and the vehicle is extracted from the frame image. Extract the part. And the recognition part 112 identifies a license plate position from the extracted vehicle part, and recognizes the vehicle information described in the said license plate by the OCR process using the template for a comparison.

  Thus, according to the present embodiment, the parameter specifying unit 115 of the vehicle information recognition apparatus 100 specifies the current parameter, so that the user can adjust the adjustment unit as indicated by the difference value between the current parameter and the target parameter. By performing this adjustment, the parameters of the vehicle information recognition device 100 can be set appropriately and easily.

  Further, according to the present embodiment, the height and the angle of view are adjusted so that the rectangular area can be accommodated in the moving image in step S2, and after the direction angle and the depression angle are within the target parameter error range in step S6, step S8 is performed. Use to adjust the angle of view. Thereby, when one of the lane boundary lines is not captured in the captured image when the lane is imaged with the target parameter, the direction angle, the depression angle, and the angle of view can be adjusted.

As described above, the embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to the above, and various design changes and the like can be made without departing from the scope of the present invention. It is possible to
For example, in the present embodiment, a case has been described in which the parameters are adjusted after adjusting the height and the angle of view so that the rectangular area is included in the moving image in step S2, but the present invention is not limited to this. For example, even when a part of the rectangular area does not fit in the moving image in step S2, the same processing can be performed when the I / F unit 113 accepts input of coordinates outside the moving image. As a method for the I / F unit 113 to accept input of coordinates outside the moving image, for example, a method of reducing a captured image and displaying it on a touch panel can be cited.

  In this embodiment, the case where the parameters to be adjusted are the depression angle, the direction angle, the angle of view, and the height has been described. However, the present invention is not limited to this, and various setting items on the camera side for adjusting the captured image. If so, other parameters may be adjusted. Examples of parameters are the imaging direction on the horizontal plane (left and right shooting direction), the imaging direction on the vertical plane (up and down shooting direction), the angle of view, the camera installation position, the focal length, the focus state, the aperture state, Examples include camera shutter speed, camera sensitivity, and optical filter characteristics.

  In the present embodiment, the case where the presenting unit 117 presents the difference value between the current parameter and the target parameter has been described. However, the present invention is not limited to this. For example, the present parameter and the target parameter may be presented side by side. . Only the current parameters may be presented. In addition, the presentation unit 117 may present the comparison result between the current parameter and the target parameter by using an arrow or a sentence to indicate the direction or amount to be adjusted.

  In the present embodiment, the case where a comparison template that is a character image cut out from a license plate image captured with the target parameter is used as the setting information used by the recognition unit 112 to recognize vehicle information has been described. It is not limited to this. For example, it may be a value indicating the character size when the license plate is imaged with the target parameter, or may be OCR setting information that has been machine-learned using a plurality of comparison templates.

  In the present embodiment, the case where the coordinate specifying unit 114 specifies the two-dimensional coordinates of the four points in the rectangular area has been described. However, the present invention is not limited to this, and three-dimensional coordinates may be specified. Specifically, the I / F unit 113 displays rectangular three-dimensional CAD data, and the user rotates, moves, and enlarges / reduces the CAD data so that it matches the rectangular area in the captured image. The specifying unit 114 can specify the three-dimensional coordinates of the vertex.

  Moreover, although this embodiment demonstrated the case where the imaging device by this invention was mounted in the vehicle information recognition apparatus 100, it is not restricted to this, For example, the imaging device which requires installation with a predetermined parameter, such as an imaging device for surveying Even if applied to the above, the same effect can be obtained.

  The operation of each processing unit described above is stored in a computer-readable recording medium in the form of a program. Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Alternatively, the computer program may be distributed to the computer via a communication line, and the computer that has received the distribution may execute the program.

  The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, what is called a difference file (difference program) may be sufficient.

  DESCRIPTION OF SYMBOLS 100 ... Vehicle information recognition apparatus 110 ... Camera 111 ... Imaging part 112 ... Recognition part 113 ... I / F part 114 ... Coordinate specific part 115 ... Parameter specific part 116 ... Target parameter memory | storage part 117 ... Presentation part 120 ... Zoom lens 121 ... Zoom Mechanism 130 ... Tripod 131 ... Leg 132 ... Camera support 133 ... Horizontal rotation mechanism 134 ... Vertical rotation mechanism 135 ... Vertical extension / contraction mechanism

Claims (8)

An adjustment method for an imaging apparatus,
An imaging step in which an imaging unit of the imaging device captures an image including a part of a rectangular region on a road surface;
A coordinate specifying step in which the coordinate specifying unit of the imaging device specifies the coordinates of each vertex of the rectangular region based on the image captured by the imaging unit;
A parameter specifying step of specifying the parameter based on the shape of a closed section connecting the coordinates specified by the coordinate specifying unit;
An adjustment method for an imaging apparatus, comprising: an adjustment step of adjusting an adjustment unit of the imaging apparatus so that the parameter specified by the parameter identification section is referred to and the parameter is brought close to a target parameter. The method for adjusting an imaging apparatus according to claim 1, wherein the rectangular area has a line at both ends of a lane and a line orthogonal to the line. Before the imaging step, the image capturing unit has an image widening step of adjusting the angle of view by the adjustment unit of the imaging device so that the image captured by the imaging unit includes all of the rectangular area,
3. The adjustment according to claim 1, wherein, in the adjustment step, after adjusting the direction angle and the depression angle among the parameters to be close to the target parameter, the adjustment is performed so that the angle of view is close to the target parameter. Adjustment method of imaging apparatus. An imaging unit that captures an image including a part of a rectangular area on the road surface;
An adjustment unit for adjusting a parameter for determining an imaging range of the imaging unit;
A coordinate identifying unit that identifies the coordinates of each vertex of the rectangular region based on the image captured by the imaging unit;
An image pickup apparatus comprising: a parameter specifying unit that specifies the parameter based on a shape of a closed section connecting coordinates specified by the coordinate specifying unit. A recognition unit for recognizing vehicle information from an image captured by the imaging unit using setting information designed to achieve the best recognition accuracy when the imaging range determined by the target parameter is captured. The imaging device according to claim 4. The imaging device according to claim 4, wherein the rectangular area has a line at both ends of a lane and a line orthogonal to the line as a side. The imaging apparatus according to claim 4, further comprising: a presentation unit that compares the parameter specified by the parameter specifying unit with a predetermined target parameter and presents the comparison result. . The imaging device
An imaging unit that captures an image including a part of a rectangular area on the road surface;
A coordinate identifying unit that identifies the coordinates of each vertex of the rectangular region based on the image captured by the imaging unit;
A program for functioning as a parameter specifying unit that specifies the parameter based on a shape of a closed section connecting coordinates specified by the coordinate specifying unit.

Images