JP3427619B2 - Overhead wire imaging apparatus and method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an overhead wire photographing apparatus and method, and more specifically, to an overhead power line and an overhead wire such as a ground wire for preventing lightning strikes thereof.
The present invention relates to an overhead wire photographing apparatus and method for photographing in an installed state or a used state.

[0002]

2. Description of the Related Art Conventionally, as a method of inspecting an overhead line, a method has been adopted in which a subject overhead line is video-recorded while flying at a very low speed by a helicopter beside the overhead line, and then returned to the ground to be reproduced and visually inspected. Has been. When shooting in the air, the first
To be properly focused on the target transmission line, secondly, the transmission line should always be located in the center of the screen, and thirdly, the flow of the image so that the details of the transmission line can be confirmed on the playback screen. It is necessary to satisfy the condition that there are few.

In the conventional example, in order to satisfy these conditions, an operator who rides on a helicopter manually operates a video camera so that the object is positioned substantially in the center of the screen while focusing on the object. It was Such camera operation requires considerable skill.

[0004]

However, the method of manually operating the video camera is not very efficient, and cannot obtain sufficient image information. For example, in order to be able to check the details of the power transmission line (welding points due to broken wires, lightning, etc.), it is necessary to increase the shooting magnification as much as possible. It becomes extremely difficult to operate the camera to focus on the image and hold it in the center of the screen. If helicopter flight is added to this, the difficulty of camera operation will be compounded, and in particular focus adjustment will be almost impossible. Therefore, in the conventional example,
The flight speed of the helicopter was at most 2-3 km / h.

In order to reduce the flow of images, it is effective to increase the shutter speed and reduce the flight speed of the helicopter or hover. But,
Generally, when you slow down or hover,
The length of the power transmission line that can be photographed in one flight is shortened, which is reflected in the cost.

Most consumer video cameras use auto
It has a focus function, and it seems that at least the focus adjustment can be automated by using this, but it is also a distance measuring area within the screen (usually a narrow area in the center of the screen) Need to be located. That is, unless the function of automatically maintaining the transmission line or the like in the center of the screen is put into practical use, even if only the focus adjustment is automated, it cannot be used.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the above problems and provide an overhead line radiographing apparatus and method in which camera operation is automated as much as possible.

It is another object of the present invention to provide an overhead line radiographing apparatus and method capable of reliably photographing an object even at a higher speed.

[0009]

According to the present invention, a measurement value of a distance / direction measuring means capable of precisely measuring a distance and a direction to an overhead line, and a photographing angle for changing a focus of the image pickup means and a photographing angle of the image pickup means. The photographing direction angle by the changing means is calibrated in advance. The measurement result of the distance / direction measuring means is calibrated by the calibration data obtained thereby, and the focus of the image pickup means and the photographing direction are automatically controlled. Thereby, the overhead line can be automatically captured in a focused state within the photographing screen of the image pickup means.

Image processing of the image taken by the image pickup means is carried out to detect the position of the overhead line in the shooting screen, and based on the result,
The shooting direction of the image pickup means is controlled so that the overhead line is located at a predetermined position within the shooting screen. As a result, the overhead line can be positioned at a predetermined position on the photographing screen, for example, at a substantially horizontal position in the center. This facilitates detailed observation of the overhead line on the monitor screen or the reproduction screen.

By setting the distance / direction measuring means so as to output the measured value of the closest distance within a preset distance range, for example, a wire similar to an overhead wire or a plurality of overhead When there is a line, the distance and direction of one overhead line of interest can be measured appropriately. By using the laser measuring means, the direction and distance can be measured at high speed and with high accuracy.

By activating the shooting direction control by image processing in response to the normal recognition of the overhead line by the distance / direction measuring means, for example, when the overhead line is completely out of the visual field of the imaging means, the distance / direction measurement is performed. By controlling the shooting direction based on the measurement result of the means, the overhead line can be quickly brought into the field of view.

As the image processing of the photographed image, a plurality of lines in a predetermined direction are extracted from the photographed image by the image pickup means, the center coordinates of the overhead line are calculated from each line, and the linear regression equation is obtained from the obtained center coordinate data. By calculating, the position and direction (inclination) of the overhead line can be calculated by a simple calculation,
Real-time control can be realized.

The success / failure of the overhead line recognition can be easily determined by using the correlation coefficient of a plurality of central coordinate data of the overhead line.

By providing the recording means for recording the photographed image on the recording medium, the overhead line can be observed in detail afterwards. By providing the video display means for displaying the shot image as a video, the overhead line can be confirmed in detail while shooting.

By providing the operating means for manually operating the focus and the photographing direction of the image pickup means, the accuracy of the corresponding data or the calibration data can be improved, and the manual adjustment can be performed as necessary.

Particularly in the method according to the present invention, the calibration data is applied to the distance value and the direction value measured by the distance / direction measuring means in the search step to automatically control the focus and the photographing direction of the image pickup means. By doing
Automatically, the overhead line can be almost certainly put in the field of view in focus. Further, in the tracking step after the search step, while automatically controlling the focus of the image pickup means according to the distance value measured by the distance / direction measuring means, the position in the image plane of the overhead line from the image taken by the image pickup means is automatically controlled. Is calculated and the shooting direction and angle of the imaging means are controlled so that the overhead line is located at a predetermined position in the shooting screen, so that the overhead line is at a predetermined position in the shooting screen (for example, a substantially horizontal position in the center of the screen). Since it is located at, it becomes easier to confirm the details on the spot or on the reproduction screen after the fact.

The calibration data can be made more precise by providing an automatic calibration step in addition to the manual calibration step in the calibration step. By executing the automatic calibration step each time photographing is performed, the calibration data can be updated to the content that immediately responds to the change in the situation for each photographing, and the accuracy of the search step becomes higher.

In the automatic calibration step, first, the calibration data regarding the shooting direction is corrected, and then the calibration data regarding the distance is corrected. By doing this, it becomes easier to collect the calibration data, and it becomes easier to ensure the required accuracy.

[0020]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a schematic block diagram of an embodiment of the present invention. Reference numeral 10 is a video camera for photographing an overhead line, and a camera mount 1 for preventing vibration by a helicopter.
It is fixed at 2. The camera gantry control device 14 tilts the camera gantry 12 (that is, the camera 10) in a tilt (up and down) direction and a roll direction (when the camera 10 is viewed from the front, according to control signals from the camera gantry operation console 16 and the computer 18). Direction or counterclockwise rotation). The current tilt angle and roll angle of the camera mount 12 are constantly supplied to the camera mount controller 14. The focus of the taking lens of the video camera 10 can be externally controlled, and the current focus value is supplied to the camera mount controller 14. Camera stand controller 14
Supplies the angle data from the camera mount 12 and the focus data from the video camera 10 to the computer 18 at all times or in response to a request from the computer 18.

In this embodiment, the video camera 10 is used.
A high-definition camera was used for. Although the taking lens of the camera 10 is a zoom lens, it is fixed at 12 times or 24 times and used when taking an overhead line.

The output video signal of the video camera 10 is applied to the image processing device 20 via the camera mount controller 14. The image processing device 20 is a digital arithmetic device using a digital signal processor (DSP), and the position of the power transmission line on the photographing screen by image processing of the output video signal of the video camera 10 will be described later in detail. And the inclination and the like are calculated and supplied to the computer 18.

Reference numeral 22 denotes a laser scanner (or laser range finder) for precisely measuring the direction and distance of the object by the direction and time when the irradiated laser pulse hits the object and returns. The result is computer 18
Is supplied to. The laser scanner 22 of the present embodiment is the same as that of the DR. Johannes Riegl
It is MS-Q140 manufactured by the same company. Measuring range is 10-5
0m, data rate 12kHz, scanning angle ± 30 °, scan rate 6Hz (3 round trips), beam size 3mrad (15cmφ at 50m), interface ECP parallel, measurement accuracy ± 10c.
m. In this embodiment, the distance is set to be measured every 0.03 °.

In addition to the computer 18, operation keys 24 for inputting various instructions and data, and a monitor (liquid crystal display device) 26 for displaying the operating status of the computer 18 are connected. Various kinds of software described below, such as automatic tracking software, are installed in the computer 18.

Reference numeral 28 is a video tape recorder for recording the output video signal of the camera 10, and is a computer 1
The measurement position and time information supplied from 8 is also recorded on the video tape at the same time so as to be superimposed on the video signal or separable from the video signal. The photographing position is, for example, GPS (G
(Global Positioning System)
Depending on the method, it can be easily measured with high accuracy. Also, 29
Is a high-definition monitor for displaying an output video signal of the camera 10. The monitor 29 is used for the manual adjustment of the focus of the camera 10, the confirmation of the automatic adjustment result, and the direct observation of the taken overhead line.

FIG. 2 shows a flow chart of the main routine of this embodiment. Immediately after the power is turned on, a manual mode in which the orientation and focus of the camera 10 can be manually operated is set (S
1). In the manual mode, the operator can manually adjust the orientation and focus of the camera 10 by operating the joystick for operating the shooting direction and the focus knob for focus adjustment of the camera platform console 16.

When the operator turns on the automatic tracking start button of the operation key 24 (S2), the initialization mode is set (S3). In the initialization mode, first, the camera mount 12 is controlled to direct the camera 10 in an initial direction (for example, a horizontal 90 ° direction with respect to the machine body), and the angle information and direction control of the camera mount 12 are initialized. Next, if necessary, the camera 10
Manually check the correspondence between the focus value and the direction of the camera mount 12 and the measurement distance and direction of the laser scanner 22,
Save as calibration data (manual or coarse calibration). Immediately after mounting this device on the helicopter, the relationship between the mounting angle of the laser scanner 22 and the tilt angle of the camera mount 12, and the measurement distance of the laser scanner 22 and the camera 1 are described.
This is because it is necessary to confirm the correspondence with the focus value of 0 at least once. After the initialization of the orientation of the camera mount 12 or the manual calibration, the calibration data obtained before or the calibration data obtained by the previous manual calibration is used as the image processing device 20.
Compensate using (automatic calibration or precision calibration).

A method of correcting the calibration data will be briefly described.
If the angle measured by the laser scanner 22 is R, the tilt angle of the camera mount 12 is C, and the calibration value is D, these generally satisfy the following formula C = R + D. The tilt angle of the camera mount 12 is controlled by the previous calibration value D, and the position of the electric wire is calculated by the image processing of the captured image at the tilt angle by the image processing device 20. A specific method of calculating the position of the overhead line will be described later. The vertical angle of view of the overhead line on the shooting screen is W, the length of the imaging surface is V,
When the focal length is f, W / 2 = arctan (V / S · f).

When the electric wire is deviated from the center of the photographing screen, the correction amount Δd of the calibration data is Δd = P · (W / L), where P is the number of pixels of the electric wire deviated from the center of the photographing screen, L
Is the number of pixels in the vertical direction (vertical direction) of the shooting screen. When the corrected calibration data is Dn and the uncorrected calibration data is Do, Dn = Do + Δd.

The relationship between the measurement distance of the laser scanner 22 and the focus value of the camera 10 is as follows. First,
The relationship between the focus value output from the camera 10 and the distance is accurately measured on the ground. The distance value converted from the focus value is called the focus distance. Laser scanner 2
When the measurement distance of 2 is R, the focus distance is F, and the calibration value is D, F = R + D.

The focus of the camera 10 is controlled by the previous calibration value D. The focus distance is moved forward while the captured image is being processed by the image processing apparatus 20, and the measurement distance of the laser scanner 22 when the electric wire cannot be recognized due to the image processing of the image processing apparatus 20 becomes R.
1. The focus distance is shifted backward, and the measurement distance of the laser scanner 22 when the electric wire cannot be recognized by the image processing of the image processing apparatus 20 is R2. The corrected calibration Dn is given by the following equation. That is, Dn = F- (R1 + R2) / 2 The correction of the calibration value generally includes replacement with a new measurement result and modification of the old calibration value with a new measurement result.
As a matter of fact, no subtle differences have been found so far regardless of which one is adopted. In the above example of the correction method, the angle is modified and the distance is replaced.

The laser scanner 22 is only required for manual calibration.
Between the mounting angle of the camera and the tilt angle of the camera mount 12,
Also, regarding the relationship between the measurement distance of the laser scanner 22 and the focus value of the camera 10, only rough correspondence that relies on human vision is known, and the accuracy is poor. In this embodiment, the more accurate correspondence can be known by correcting the result of the manual calibration by the automatic calibration. As a result, the photographing magnification can be increased, and the helicopter can fly at higher speed. Details of manual calibration and automatic calibration will be described later.

When accurate calibration data corrected by the automatic calibration is obtained, the search mode is set (S4). In the search mode, the camera 10 and the camera mount 12 are automatically controlled according to the measurement result of the laser scanner 22. That is, when an electric wire enters the scan range, the computer 18
Applies calibration data to the measurement values (direction and distance) by the laser scanner 22 to obtain the direction and focus value of the camera 10, and controls the tilt angle of the camera mount 12 and the focus of the camera 10. This allows the camera 1
0 is almost in focus on the overhead line of the shooting target, and
It can be located almost in the center of the screen.

When the object to be photographed can be detected in the search mode, the track mode is set (S5). In the track mode, the image processing apparatus 20 obtains the position and the inclination of the electric wire in the photographing screen, and controls the tilt angle and the roll angle of the camera mount 12 based on the result so that the electric wire becomes almost horizontal in the center of the screen. To Further, the focus of the camera 10 is automatically adjusted by the measurement distance of the laser scanner 22 and the corrected calibration data. When the electric wire cannot be recognized due to sudden movement of the electric wire in the image or out-of-focus due to a focus defect (S6), the search mode is returned to and the electric wire is detected again (S4).

When the operator pushes the automatic tracking end button of the operation key 24 (S7), the automatic tracking is ended and the manual mode is set (S8).

The computer 18 starts the operation of the VTR 28 in the recording mode by turning on the automatic tracking start button, and stops the recording operation of the VTR 28 by turning on the automatic tracking end button. As a result, the output video signal of the camera 10 between S2 and S7 and the information of the shooting position and time are recorded on the video tape. The recording start and end of the VTR 28 may be linked with another operation, or may be operated individually by the operator.

FIG. 3 shows a detailed flowchart of the manual calibration in the initialization mode (S3) of FIG.

The pilot hovers at a normal photographing position (a position distant from the electric wire to be photographed by a certain distance) or flies the helicopter at a low speed along the electric wire (S1).
1). At that time, the operator manually adjusts the shooting direction and focus of the camera 10 so that the overhead line is clearly displayed in the shooting screen of the camera 10 in the normal shooting state (S).
12). At the same time, the laser scanner 22 is operated to confirm that the overhead line of the object is recognized (S13). When the object is clearly displayed in the video screen and the laser scanner 22 can measure the same object (S14), the manual calibration button of the operation key 24 is pressed (S).
15). Repeat this several times. By this series of operations, a relatively rough correspondence between the measurement distance of the laser scanner 22 and the focus value of the camera 10, and the laser
Data (calibration data) showing a relatively rough correspondence between the measurement angle of the scanner 22 and the direction of the camera mount 12 can be collected. The obtained calibration data (an average value of the obtained data when repeated several times) is stored in the memory (hard disk or the like) of the computer 18 (S1).
6).

Detailed flow charts of the automatic calibration in the initialization mode (S3) of FIG. 2 are shown in FIGS. The automatic calibration is a process of generating more accurate calibration data by using the calibration data immediately before the manual calibration or the calibration data corrected by the automatic calibration at the previous photographing as the initial data. Further, FIG. 6 is a diagram showing the relationship between the shooting direction of the camera 10, the focus, the scan range of the laser scanner 22, and the electric wires to be shot. In FIG. 6, the electric wire 30
Extends perpendicularly to the paper surface, and the helicopter flies along the electric wire 30 in a direction perpendicular to the paper surface while keeping the distance from the electric wire 30 as constant as possible. The pilot flies at the normal photographing position and speed (S21), and operates the laser scanner 22 to recognize the electric wire to be photographed. According to the initial calibration data and the measurement value of the laser scanner 22,
The focus of the camera 10 and the direction of the camera mount 12 are automatically controlled (S22, S23). As a result, the electric wire to be photographed should be positioned substantially in the center of the photographing screen of the camera 10 in a substantially focused state.

The electric wire to be photographed is recognized by the image processing device 20 (S25, S26). Details of the electric wire recognition process by the image processing device 20 will be described later. When the electric wire to be photographed is not within the photographing screen of the camera 10, that is,
When the electric wire cannot be recognized by the image processing device 20 (S26), the operator auxiliary operates the tilt fine adjustment knob of the camera pedestal console 16 to finely adjust the tilt angle of the camera pedestal 12 to obtain the object to be photographed. The electric wire is the camera 10.
(S27).

When the electric wire is recognized by the image processing apparatus 20 (S26), the tilt angle control of the camera mount 12 is shifted from the control based on the measured value of the laser scanner 22 to the control based on the electric wire recognition result of the image processing apparatus 20. . That is, according to the processing result of the image processing device 20, the electric wire is connected to the camera 10.
The tilt angle of the camera mount 12 is controlled so as to be located at the center of the shooting screen (S28 to S33).

First, the data processing number and counter for averaging processing are set (S28). It is obvious that the number of data processing may be set in advance. The laser scanner 22 can recognize the electric wire (S29), and the image processing device 20 can recognize the electric wire (S3).
0) is confirmed, the tilt angle of the camera mount 12 is read (S31), and the tilt angle of the camera mount 12 is controlled so that the electric wire is located at the center of the screen according to the electric wire recognition result by the image processing device 20. (S32), the correspondence between the tilt angle of the controlled camera mount 12 and the measurement direction of the laser scanner 22 is stored.

After repeating S29 to S32 the number of times of data processing set in S28 (S33), the recorded data is averaged (S34). Initial data is corrected by the average value (S35). Alternatively, the correspondence data obtained in S34 is used as the calibration data indicating the correspondence between the measurement direction of the laser scanner 22 and the tilt angle of the camera mount 12.

After the angle is calibrated, the correspondence between the focus of the camera 10 and the measurement distance of the laser scanner 22 is calibrated (S36 to S48). Specifically, after confirming that the laser scanner 22 can recognize the electric wire within the scanner range (S36), the measurement direction is applied to the calibration data determined in S35 to apply the tilt angle of the camera mount 12. Is controlled (S37). As a result, the electric wire to be photographed is located substantially in the center of the photographing screen of the camera 10.

At this stage, the computer 18 has the camera 1
The focus of 0 is finely adjusted to move the focus of the camera 10 to the front side of the electric wire (S38), and it is checked whether or not the electric wire can be recognized by the image processing apparatus 20 in association with the operation (S39). The limit point of the electric wire recognition by the image processing device 20 is detected (S40), and the focus value of the camera 10 at that time is read and temporarily stored (S41). This time, the focus of the camera 10 is moved to the far side (S4
2) Read the focus value of the camera 10 at the limit point of electric wire recognition by the image processing device 20 (S43, S).
44, S45).

After confirming that the laser scanner 22 can recognize the electric wire (S46), the average value of the focus value read in S41 and the focus value read in S45 is calculated (S47). The initial calibration data is corrected by the measurement distance obtained by the laser scanner 46 at S46 and the average focus value calculated at S47 (S4).
8). Here again, the correspondence between the measurement distance obtained by the laser scanner 22 in S46 and the average focus value calculated in S47 may be used as new calibration data.

A problem with distance calibration is the relationship between the distance measurement accuracy of the laser scanner 22 and the depth of field. In this embodiment, a 2/3 inch high-definition camera is used as the camera 10, and a 12 × to 24 × zoom lens is used as the lens of the camera 10. The size D of the subject that can be actually photographed is D = d × L / f, where L is the distance from the camera 10, f is the focal length of the camera, and d is the size of the imaging surface of the camera 10. is there. When the focal length f is 12 times, it is 102 mm,
It is 204 mm when multiplied by 24. FIG. 7 shows a table of the relationship between the shooting range and the resolution with respect to the distance L. At a distance of 100 m, it is possible to photograph a range of a maximum width of 9.6 m and a vertical width of 5.3 m.

Similarly, with a 2 / 3-inch high-definition camera, maximum zoom (f = 204 mm), aperture open (F = 1.2), minimum circle of confusion δ = 0.04 mm (for example, 10-inch monitor). The relationship between the distance and the depth of field is shown in FIG.

For example, for transmission voltages of 77 kV, 270 kV and 500 kV, the ground wire diameters are 7.8 mm, 17.5 mm and 17.5 mm, respectively, and the accessible distances are 15 m, 30 m and 30 m. Therefore, the closest thing you can get when shooting electric wires is the transmission voltage of 77k.
In the case of V, it is 15 m, and the depth of field at this distance is shown in FIG.
To 52 cm. On the other hand, the distance accuracy of the laser scanner 22 is ± 10 cm as described above. Therefore, even if the focus of the camera 10 is controlled by the measurement distance data of the laser scanner 22,
Since the measurement accuracy of 2 is within the depth of field, there is no problem in this embodiment.

Since the automatic calibration shown in FIGS. 4 and 5 is automatically completed in a short time, it is preferable that the automatic calibration is set to be performed before the start of the photographing work as in this embodiment. Focus near side movement (S38) and far side movement (S42)
With regard to the above, although the operator may operate the focus fine adjustment knob provided on the camera pedestal console 16, the automatic adjustment by the computer 18 is of course quicker.

FIG. 9 is a flowchart of electric wire recognition by the laser scanner 22. The distance range to be measured is set (S51), the laser pulse is scanned within the scan range, and the reflection time (that is, distance) of the reflected light data is stored in an array according to the irradiation angle (S52). From the array data, the shortest data within the distance range set in S51 is extracted (S53). If the measurement data does not exist within the set distance range (S54), it means that the electric wire recognition has failed and an error is returned.

Even if there is data within the set distance (S5
4) If the distance or direction is too different from the immediately previous data, it is also abnormal data. Therefore, when there is immediately preceding data (S55), the distance and direction of the immediately preceding data are compared (S56), and if they are too different, it is judged as abnormal data (S57) and an abnormality is returned.

If there is no immediately preceding data (S55), or
If both the distance and the direction are close to each other even if there is the immediately preceding data (S57), the reflection time information (or distance information) of the data recognized as the shortest and the array number (or the angle information obtained from the array number) are immediately before. Stored as data (S58),
It returns with the distance and its angle determined to be the shortest within the set distance range as the return value.

Next, the recognition of electric wires by the image processing apparatus 20 will be described in detail. The image processing device 20 extracts only the outer periphery of the electric wire by edge detection and binarization processing, and calculates a linear regression equation from the image coordinate values by the least square method,
As shown in FIG. 10, the shift amount and the tilt from the center of the shooting screen of the camera 10 are calculated. At the same time, whether or not the extracted image is an electric wire is determined by a correlation coefficient. For the image processing device 20, for example, a digital signal processor (DSP) is used.

Many techniques have been proposed so far for detecting edges and lines. Among them, the method based on the most basic idea is spatial differential processing using a difference operator. Roberts, Prewit as first derivative
Well-known are t, Sobel, Laplacian of second derivative, and the like. Further, using eight differential masks, the edge strength and direction are obtained from the maximum output value, P
Template types using operators such as rewitt, Kirsch, and Robinson are also common.

The template type edge extraction method using the Robinson operator will be briefly described below. This is a method of detecting edge elements by preparing a plurality of templates assuming a gray pattern near edges on an image and calculating a correlation between the images. In practice, eight 3 × 3 templates corresponding to edges in eight directions are often used. A Robinson edge detection operator or template is shown in FIG. Direction from dark to light on the edge (arrow direction)
There are eight types as shown in FIG. (1) of FIG.
(8) to (8) correspond to the directions shown in (1) to (8) of FIG. 12, respectively.

A local product-sum operation is performed by applying each template to the target image. Then, the direction of the template in which the maximum output is obtained is the gradient direction, and the output value at that time is the gradient intensity. Since this method uses a large number of directional templates, it is possible to extract information about the direction of the edge more accurately than the differential operator. However, as will be described later in detail, in the present embodiment, the template is also (−2, 0, 2) or (−) because the one-dimensional processing is used instead of the two-dimensional processing in order to simplify the calculation.
1, 0, 1) may be one-dimensional.

FIG. 13 shows an operation flowchart of the image processing apparatus 20, and FIG. 14 shows a schematic diagram of the electric wire recognition processing. The output video signal of the camera 10 is applied to the image processing device 20 via the camera pedestal control device 14, and the image processing device 20 converts the input video signal (only the brightness signal is required) into a digital signal. The image data for one screen is stored in the internal image memory (S61). The number of extraction steps previously set in the computer 18 is set in the counter (S62). The number of extraction steps determines the number of vertical lines to be extracted within one screen. For example, in FIG.
4, x1, x2, ...
xn vertical lines are extracted from one screen.

Pixel data for one line in the vertical direction is extracted (S63), and the line data is (-2,0,2).
The template is applied to calculate the sum of products, and the absolute value is calculated (S64). From the calculation result for one line in S64, the average value and standard deviation of the data of that line are obtained (S6
5) Using the obtained average value and standard deviation, the calculation result for one line in S64 is binarized (S66). The binarized data of one line is searched from both outer sides to detect the outer edge of the electric wire, and the middle position of the two outer edge positions is set as the center of the electric wire (S67). In this way, the center positions y1 to yn of the electric wires can be calculated for n vertical lines.

Center position y of the wire for n vertical lines
When the calculation of 1 to yn is completed (S68), the coefficients a and b and the correlation coefficient r of the linear regression line y = a + bx by the least square method are calculated from the obtained coordinate values (x1, y1) to (xn, yn). Calculate by the following formula (S
69).

[0062]

[Equation 1]

Based on the obtained correlation coefficient r, the quality of the correlation is judged (S70). If the correlation is good (S70), it means that the electric wire has been photographed correctly, the inclination of the electric wire and the deviation amount from the center are calculated from the primary regression line (S71), and the calculated value (deviation amount and inclination) is displayed in the computer 18. ) Is notified.
If the correlation is bad (S70), the recognition of the electric wire has failed, so a code indicating mistrack is substituted for the return value (S72), and the computer 18 is notified of abnormal termination.

In the image processing apparatus 20 of this embodiment, the edge is detected by the one-dimensional processing, so that the amount of calculation and the time can be greatly reduced as compared with the case of the two-dimensional processing. As a result, the rotation control (tilt direction and roll direction) of the camera mount 12 based on the electric wire recognition result of the image processing device 20 can be performed in real time and at a relatively low cost. Moreover, since the correlation of the center positions of the electric wires in each vertical line is checked to confirm the success of the electric wire recognition, an erroneous object is not mistakenly recognized as an electric wire. Background (generally sky or forest)
Hard to be confused by.

According to the present embodiment, since the electric wire is recognized by the laser scanner and the camera is automatically controlled in the measuring distance and direction, if the scanning range of the laser scanner is satisfied,
It is possible for the pilot to reliably and quickly bring the power transmission line into the shooting range of the camera simply by causing the pilot to fly at a rough position with respect to the power line. Further, by calibrating the focus and shooting direction of the camera and the measurement distance and measurement direction of the laser scanner in advance, the focus and shooting direction of the camera can be controlled with high accuracy during actual shooting. As a result, shooting with a higher shooting magnification can be realized.

The correspondence between the measurement distance by the laser scanner and the focus value of the camera is measured in advance, and at the time of actual photographing, the measurement distance by the laser scanner is calibrated with the measurement data and used for focus control of the camera. Therefore, the precision of focus control can be further increased, and a clear image can be obtained.

Further, the deviation and the inclination of the electric wire in the photographing screen are calculated by the image processing, and the tilt angle and the roll angle of the camera are controlled so that the electric wire is positioned substantially horizontally in the center of the screen. Even if there is some disturbance in the position or the power transmission line of the shooting target, it is possible to obtain a stable image in which the electric wire is positioned in the center of the screen without any inclination.

All of the devices shown in FIG. 1 are usually mounted on a helicopter, but some of them, for example, VTR2.
8 and the like may be provided on the ground and the video signal of the camera 10 may be supplied to the VTR 28 on the ground via the wireless transmission system. Of course, an unmanned aerial vehicle may be used instead of the manned helicopter, depending on the weight and bulk of each device.

[0069]

As can be easily understood from the above description, according to the present invention, it is not necessary for the operator to perform a fine operation during photographing, and the photographing of the overhead line can be performed at higher photographing magnification and / or higher speed. You will be able to.

Since the photographing direction and the focus are controlled in real time, it is possible to realize automatic focus on the overhead line and to automatically track the overhead line. The burden on the operator is greatly reduced, and it is possible to shoot only with the pilot. Furthermore, as a result of automatic control of focus and shooting direction,
High quality images can be obtained. By increasing the flight speed, safety can be improved and cost can be reduced.

[Brief description of drawings]

FIG. 1 is a schematic block diagram of an embodiment of the present invention.

FIG. 2 is a flowchart of a main routine of this embodiment.

FIG. 3 is a detailed flowchart of manual calibration in the initialization mode (S3) of FIG.

FIG. 4 is a part of a detailed flowchart of automatic calibration in the initialization mode (S3) of FIG.

5 is a part of a detailed flowchart of automatic calibration in the initialization mode (S3) of FIG.

FIG. 6 is a diagram showing a relationship between a camera 10, a laser scanner 22 and an electric wire to be photographed.

FIG. 7 is a table showing a relationship between a shooting range and resolution with respect to a distance L.

FIG. 8 is an example of a relationship between distance and depth of field.

9 is a flowchart of electric wire recognition by the laser scanner 22. FIG.

FIG. 10 is an example of photographing an electric wire on a photographing screen of the camera 10.

FIG. 11 is a Robinson edge detection operator or template.

12 is a direction of an edge that can be detected by the template shown in FIG.

FIG. 13 is an operation flowchart of the image processing apparatus 20.

FIG. 14 is a schematic diagram of electric wire recognition processing in the image processing apparatus 20.

[Explanation of symbols]

10: Video camera 12: Camera mount 14: Camera stand control device 16: Camera platform operation console 18: Computer 20: Image processing device 22: Laser scanner 24: Operation key 26: Monitor 28: Video tape recorder 29: High-definition monitor 30: Electric wire

─────────────────────────────────────────────────── ─── Continuation of front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) G03B 15/00-15/02 H02G 1/02 323 H02G 15/00-15/196 H04N 5/222-5 / 257 G06T 1/00

Claims (14)

(57) [Claims] 1. An overhead line photographing apparatus for photographing an overhead line, comprising: an image pickup means for converting an optical image into an electric signal; a photographing angle changing means for changing a photographing angle of the image pickup means; Distance to and distance to measure direction
Direction measuring means, distance calibration data indicating the correspondence between the distance value measured by the distance / direction measuring means and the focus value of the image capturing means, and the direction measured by the distance / direction measuring means and the photographing angle. A storage unit that stores direction calibration data indicating the correspondence with the shooting direction by the changing unit, and a distance value measured by the distance / direction measuring unit is calibrated by the distance calibration data to automatically control the focus of the imaging unit. The focus control means and the direction measured by the distance / direction measuring means are calibrated by the direction calibration data to control the photographing angle changing means, and thus the image pickup means is automatically adjusted in the direction of photographing the overhead line. An imaginary line image is extracted from an image captured by the image capturing direction control unit and the image capturing unit,
An image processing unit that calculates the position and tilt in the shooting screen, and the shooting angle is changed according to the calculation result of the image processing unit so that the overhead line of the shooting target is located at a predetermined position in the shooting screen of the imaging unit. An overhead line radiographing apparatus, comprising: an angle control means for controlling the means. 2. The overhead line radiographing apparatus according to claim 1, wherein the distance / direction measuring means outputs a measured value of the closest distance within a preset distance range. 3. The laser measuring means according to claim 1 or 2, wherein the distance / direction measuring means is a laser measuring means that outputs laser pulsed light within a predetermined angle range and measures the distance and direction based on a reflection angle and a return time. Aerial line photography equipment. 4. The overhead line radiographing apparatus according to claim 1, further comprising control means for activating the angle control means in response to normal recognition of the overhead line by the distance / direction measuring means. . 5. The image processing means extracts a plurality of lines in a predetermined direction from an image picked up by the image pickup means, calculates center coordinates of an imaginary line from each line, and draws straight lines from the obtained plurality of center coordinate data. A regression formula is calculated.
The overhead line photographing apparatus according to any one of 1. 6. The image processing means according to claim 1, further determining success / failure of overhead line recognition from correlation coefficients of a plurality of center coordinate data of the overhead line. Overhead radiography device. 7. The overhead line radiographing apparatus according to claim 1, further comprising a recording unit that records the image captured by the image capturing unit on a recording medium. 8. The overhead line radiographing apparatus according to claim 1, further comprising video display means for video-displaying an image taken by the imaging means. 9. The overhead line radiographing apparatus according to claim 1, further comprising operation means for manually operating the focus of the photographing angle changing means and the image pickup means. 10. The overhead radiography apparatus according to claim 1, which is mounted on an aircraft. 11. An aerial line photographing method for photographing an aerial line, the method comprising: measuring a distance and a direction to an aerial line to be photographed.
Correspondence between the distance measured by the direction measuring means and the focus value of the image pickup means for converting an optical image into an electric signal, and correspondence between the direction measured by the distance / direction measuring means and the photographing direction of the image pickup means. A calibration step of measuring the calibration data shown in advance, and a search step of automatically controlling the focus and the shooting direction of the imaging means by calibrating the distance value and the direction value measured by the distance / direction measuring means with the calibration data. After placing the overhead line of the object to be imaged in the imaging field of view of the imaging means by the search step, the imaging of the imaging means while automatically controlling the focus of the imaging means by the distance value measured by the distance / direction measuring means. Calculate the position of the overhead line in the shooting screen from the captured image of the means, and place the overhead line at a predetermined position in the shooting screen. And a tracking step for controlling a photographing direction and an angle of the image pickup means, and if the imaginary line is lost in the photographing field of view of the image pickup means in the tracking step, the search step is returned. How to take a line. 12. The calibration step comprises a manual calibration step of manually adjusting a focus and a shooting direction of the imaging means in a normal recognition state of the overhead line by the distance / direction measuring means, calibration data by the manual calibration step and a previous calibration step. Using any of the calibration data at the time of shooting,
An automatic adjustment of the focus and the shooting direction of the image pickup means by the measurement value of the direction measurement means, and then an automatic adjustment of the focus and the shooting direction by the image processing of the image shot by the image pickup means to correct the calibration data used. A calibration step is provided, and the automatic calibration step uses either the calibration data obtained by the manual calibration step or the calibration data obtained at the previous photographing, and the measurement value of the distance / direction measuring means is used to measure the image capturing means. Preparatory steps for automatically adjusting the focus and the shooting direction, and finely adjusting the shooting direction so that the overhead line is located at the center of the shooting screen by image processing of the shot image by the imaging means,
A shooting direction calibration value correction step of correcting the calibration data relating to the shooting direction based on the correspondence between the resulting shooting direction and the measurement direction of the distance / direction measuring means at that time, and after the shooting direction calibration step, The focus is adjusted to the near side and the far side, and the limit focus value that can recognize the overhead line is detected by the image processing of the image captured by the image pickup means, and the average value of both limit focus values and the distance and direction at that time are measured. The overhead line radiographing method according to claim 11, comprising a distance calibration value correction step of correcting calibration data relating to the distance in correspondence with the measured distance of the means. 13. The photographing direction calibration value correction step comprises:
Correspondence between the shooting direction after finely adjusting the shooting direction so that the overhead line is located at the center of the shooting screen by image processing of the shot image by the image pickup means and the measurement direction of the distance / direction measuring means at that time point. 13. The calibration data regarding the shooting direction, and the distance calibration correction step uses the correspondence between the average value of the both limit focus values and the measured distance of the distance / direction measuring means at that time as the calibration data regarding the distance.
The overhead line photography method described in. 14. The overhead line radiographing method according to claim 12, wherein the manual calibration step is performed at least once after installation, and the automatic calibration step is performed each time an image is captured.