JP2006121261A - Method for correcting target position of camera stabilizer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for correcting a target position of a camera stabilizer in which the target position can be corrected to capture a target continuously within the field of view of a camera even if the precision of a target position acquired from a map, or the like, is low when an aircraft mounting the camera approaches the target. <P>SOLUTION: The method for correcting the target position is applied to a position control tracking system camera stabilizer where a visual axis L of the camera is directed automatically to previous target position P when the previous target position P and a current position N are given, the first time target position P1 is given from a map, its own position N is given by GPS, and the visual axis L is controlled depending on manual operation. The visual axis L is directed to the position Q of an imaging object by performing manual operation at the capturing position N1, a point closest the P1 on the visual axis L2 directing the Q from the N1 is operated as a second target position P2, and then the target position is corrected similarly as required thus correcting the target position to approach the Q gradually. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a camera stabilizer that is mounted on an aircraft body such as a helicopter and has a function of automatically tracking a camera's visual axis to a position when position information such as longitude and latitude regarding a target to be imaged is given in advance. In particular, the present invention relates to a camera stabilizer target position correction method that enables an automatic tracking target to be gradually corrected to an actual target position when the predetermined target position deviates from the actual target position.

  A camera such as a TV camera is usually mounted on a camera stabilizer in order to stably capture an object during flight from an aircraft such as a news helicopter. The camera stabilizer includes a gimbal having three degrees of freedom. The gimbal base is fixed in place on the aircraft. The gimbal is equipped with a camera and a gyro that detects the direction of the visual axis of the camera. The angle of each axis of the gimbal is detected as a gimbal angle.

  If the position of the imaging target is input as a target position from a map or the like, and the position of the aircraft is acquired by GPS (Global Positioning System), the direction (target direction) that the camera visual axis should point to can be calculated. By detecting the axial direction with a gyro, the difference between the target direction and the camera visual axis direction is calculated as a tracking error, and by constructing a control loop that minimizes the tracking error, the camera visual axis can be used even during flight of the aircraft. Position tracking that automatically points to the target position is possible.

  Japanese Patent Laid-Open No. 2001-359083 discloses an imaging apparatus that performs position tracking. This imaging apparatus includes position input means for inputting the absolute position (latitude, longitude, altitude) of the subject. The position input means is a receiving means for acquiring the absolute position (latitude, longitude, altitude) of the subject from the map and receiving the absolute position of the subject wirelessly transmitted from the keyboard or the subject itself. . In this imaging device, the absolute position of the subject is input by the position input means, the position of a moving body such as a helicopter is input by GPS, the traveling direction of the moving body is detected by the traveling direction detecting means, and the imaging direction changing unit The imaging direction of the camera is obtained, the rotation angle of the camera is controlled, and the camera visual axis is automatically tracked by the subject.

  However, the target position data that can be acquired from a map or the like is represented by latitude, longitude, and altitude, but since the accuracy of the data is generally not high, when the aircraft (moving body) approaches the target position, In many cases, an imaging target deviates from the field of view of the camera due to an error between the position of the imaging target and a target position acquired from a map or the like. In addition, the imaging target may be on the roof of a high building, but the height of the building is not usually described on the map. Therefore, when the upper position of such a high-rise building is used as the imaging target, imaging is also necessary. Due to the error between the target position and the target position acquired from a map or the like, the imaging target is likely to fall out of the camera view. In addition, since there are many maps in which altitude data is not described, altitude data may not be input, and similarly, an imaging target is likely to be out of the field of view of the camera.

  Japanese Patent Application Laid-Open No. 2001-169171 discloses a visual axis control device that can accurately measure a target position and then automatically track the camera visual axis to the target position. The visual axis control device includes: an imaging device; a control unit that controls the visual axis of the imaging device to be directed toward a target position; a visual axis controller that directs the visual axis toward the target; and a target at a plurality of points associated with movement. Computation means for computing the target position from the direction and the amount of movement. With such a configuration, the visual axis control device disclosed in Patent Document 2 captures targets whose latitude, longitude, and altitude are unknown. In order to enable automatic tracking of a target without requiring existing location data such as a map, the target is captured from at least two points with the camera, the direction of the target captured at each point, and between each point The target position data is calculated by calculation from the distance.

  Geographic information in the vicinity of the target to be imaged by the camera is represented by longitude, latitude and altitude, the geographical information is stored in advance as three-dimensional geographic data, the current position of the aircraft is acquired by GPS, and the camera and gyro are used as gimbals. A position specifying method for detecting a target position on the ground surface by mounting and stabilizing the posture of the camera with a gyro and controlling the direction of the camera (the direction of the visual axis of the camera) with the imaging control unit 35 is disclosed in Patent Document 3 ( JP-A-8-285590) is disclosed in FIG. 5 and paragraphs 0019-0022. The gimbal and gyro are equivalent to a camera stabilizer. In this position specifying method, since the position of the target is specified from the intersection of the camera visual axis when the target is viewed from the imaging position and the ground surface represented by the pre-recorded geographical information, the position of the target is determined. The three-dimensional specification is described in Patent Document 3.

  Japanese Patent Application Laid-Open No. 2004-80580 discloses another method for tracking a target so that the visual axis of a camera head is directed to the target. In this method, a moving object such as a helicopter is passed immediately above the target, the position of the moving object at that time is acquired as the position of the target, and then the visual axis of the camera pan head is set at an arbitrary position of the moving object. Aiming at the target, obtain the position, altitude and depression angle of the visual axis of the moving body at this time, calculate the altitude difference between the moving body and the target from these, and also the position, altitude and heading of the moving body Based on the angle, the azimuth and depression angles for directing the target from the moving object are calculated, and the azimuth and depression angles are used as command visual axis angles, and the visual axis of the camera platform is directed to the target based on the command visual axis angles. ing.

JP 2001-359083 A JP 2001-169171 A JP-A-8-285590 JP2004-80580

  There is a strong demand to equip a news helicopter with a camera stabilizer, and to equip the camera stabilizer with a camera so that the subject can be imaged quickly and stably by the camera. As described above, devices and methods for directing the visual axis of a camera mounted on a camera stabilizer to a target have been proposed.

  In the imaging device described in Patent Document 1, target (subject) position data is read from latitude, longitude, and altitude data from a map and input from the keyboard, or the subject's own position data is transmitted from the subject. It is received by the receiver of the imaging device. Therefore, in the imaging apparatus of Patent Document 1, whether or not the target can continue to be captured in the camera's field of view even if the camera-equipped aircraft approaches the target, that is, whether or not the proximity target can be automatically tracked is determined from the map. This depends on the accuracy of the position data of the subject to be read or the accuracy of the position data of the subject itself transmitted from the subject.

  However, since the position data of the subject that can be read from the map is quite rough, in the position tracking by the imaging device of Patent Document 1 that relies on the position data, the target object is detected by the camera when the camera-equipped aircraft approaches the target object. It is difficult to keep it in sight. In addition, since the building height data is not usually described on the map, when the target is on the roof or middle floor of the building, the altitude element in the target position data is lacking, so it is difficult to automatically track the proximity target. .

  In the visual axis control device described in Patent Document 2, a target is captured from at least two points by a camera, and the target position data is calculated from the direction of the target obtained at those points and the distance between the points. Since it is calculated, first, the vicinity of the target is reached, and until it is discovered, it depends on human observation, and after the discovery, it is necessary to capture the target with the camera at least twice. Therefore, in this visual axis control device of Patent Document 2, automatic tracking cannot be performed until the target is detected and the target is captured at least twice by the camera. Therefore, it takes time to image the target. The ability to board a plane and search for a target visually from the air depends on the skill level, so that it is difficult to stably image the target.

  In the position specifying method described in Patent Document 3, the position of the target is specified from the intersection of the camera visual axis when the target is viewed from the imaging position and the ground surface represented by the prerecorded three-dimensional geographic data. To do. Therefore, in the position specifying method of Patent Document 3, three-dimensional geographic data is essential. Further, in the position specifying method of Patent Document 3, since the target position is obtained by calculation after the target is captured by the camera, it is necessary to first discover the target and capture the target by the camera. As with the described visual axis control device, it takes time to image the target, and the ability to find the target visually in the air after boarding the aircraft depends on the skill level, so the target is imaged stably. Difficult to do.

  In the tracking method described in Patent Document 4, an azimuth angle and a depression angle directing the target from the moving body are calculated by capturing the target with the camera at a position immediately above the target and at any other position, and the azimuth And the depression angle are set as the commanded visual axis angle, and the visual axis of the camera head is directed to the target based on the commanded visual axis angle. Similarly to the visual axis control device described in Patent Document 2, It takes time until the object is imaged, and the ability to board the aircraft and search for the object visually from the air depends on the skill level, so that it is difficult to stably image the object.

  Therefore, the present invention is due to the low accuracy of the target position acquired from the map or the like, when the aircraft mounted on the camera can not keep capturing the target in the camera's field of view when approaching the target, A camera stabilizer target position correction method capable of correcting a target position is provided.

  The camera stabilizer target position correction method of the present invention enables improvement of the proximity target automatic tracking capability by the camera stabilizer even when the accuracy of the target position acquired from a map or the like is low. Further, the camera stabilizer target position correction method of the present invention has a proximity target automatic tracking capability by the camera stabilizer even when the target is on the roof or intermediate floor of a building or the like and the altitude element in the target position data is missing. improves. Furthermore, the camera stabilizer target position correction method of the present invention captures the target in the camera's field of view when the camera-equipped aircraft approaches the target because the target position acquired from a map or the like is low in accuracy. Even if it is not possible to continue and the object to be imaged cannot be directly captured visually, the camera can automatically capture the object, and thus the time from when the aircraft starts flying in the direction of the target to the start of imaging of the object is shortened. In addition, the time until the start of imaging can be made constant regardless of the skill level of the camera operator.

  In order to solve the above-mentioned problems, the present invention provides the following means.

  (1) A position control tracking type camera stabilizer that automatically points the visual axis L of the camera to the target position P when the target position P and its own position N are given. It is applied to a camera stabilizer that is provided with an operation unit for inputting an operator's operation on the map and its position N is given by GPS, and applied to the camera stabilizer in which the visual axis L is controlled according to the operation. Is directed to the target position P1, and the position Q of the imaging object is directed at the first capture position N1 where the camera can capture the position Q, thereby directing the visual axis L to the position Q of the imaging object, A point that is closest to the initial target position P1 on the visual axis L2 of the camera that is directed from the first capture position N1 to the position Q of the object to be imaged is calculated as the second target position P2, and further calculated as necessary. The visual axis L is set to the second target position P The position Q of the imaging object is directed at the second capture position N2 that can be captured by the camera, and thereby the visual axis L is directed to the position Q of the imaging object, and the second capture is performed. A point closest to the second target position P2 on the visual axis L3 of the camera that is directed from the position N2 to the position Q of the object to be imaged is calculated as the third target position P3. A camera stabilizer target position correction method for correcting the target position P so as to gradually approach the position Q of the imaging object by correcting the target position.

なる式で表し、
補正後の目標位置は、視軸Ｌ上であって、前回の目標位置に最も近い点とし、
視軸ＬのベクトルＶを法線とするとともに、前回の目標位置Ｐを通る平面Ｓを仮想し、
平面Ｓを、
Ｘｖ・（ｘ−Ｘｐ）＋Ｙｖ・（ｙ−Ｙｐ）＋Ｚｖ・（ｚ−Ｚｐ）＝０・・・・・（２）
と表し、
式（１）で表される視軸と式（２）で表される平面Ｓとの交点の座標Ｔを補正後の目標位置の座標とし、式（１）と式（２）の連立方程式を助変数ｔに関し、

と解き、式（３）及び式（１）により、座標Ｔを、
（Ｘｖ・ｔ＋Ｘｐ， Ｙｖ・ｔ＋Ｙｐ， Ｚｖ・ｔ＋Ｚｐ）・・・・・・・（４）
なる式（４）で求める
ことを特徴とするカメラスタビライザ目標位置補正方法。 (2) A position control tracking type camera stabilizer that automatically directs the visual axis L of the camera to the target position P when the target position P and the current position N are given. In a camera stabilizer target position correction method, which is applied to a camera stabilizer that includes an operation unit that inputs a visual axis L and is controlled in accordance with the operation, and corrects the target position P so as to gradually approach the position Q of the object to be imaged. ,
The coordinates of the previous target position P
P (Xp, Yp, Zp)
The coordinates N of the current position
N (Xn, Yn, Zn)
When the camera stabilizer is at the coordinate N and the camera captures the imaging object at the position Q, the vector V of the camera visual axis L is obtained.
V (Xv, Yv, Zv)
Then, by giving the auxiliary variable t, the visual axis L is

It is expressed by the formula
The corrected target position is on the visual axis L and is the closest point to the previous target position,
Assuming that the vector V of the visual axis L is a normal line, a plane S passing through the previous target position P is assumed,
Plane S,
Xv · (x−Xp) + Yv · (y−Yp) + Zv · (z−Zp) = 0 (2)
And
The coordinate T of the intersection of the visual axis represented by the equation (1) and the plane S represented by the equation (2) is set as the coordinate of the target position after correction, and the simultaneous equations of the equations (1) and (2) are expressed as follows. For the parameter t

And the coordinates T are expressed by the equations (3) and (1).
(Xv.t + Xp, Yv.t + Yp, Zv.t + Zp) (4)
A camera stabilizer target position correction method characterized by being obtained by the following equation (4).

  According to the above-described present invention, when the target position acquired from the map or the like is low, the target object cannot be continuously captured in the camera field of view when the camera-equipped aircraft approaches the target object. In addition, a camera stabilizer target position correction method capable of correcting the target position can be provided.

  The camera stabilizer target position correction method of the present invention enables improvement of the proximity target automatic tracking capability by the camera stabilizer even when the accuracy of the target position acquired from a map or the like is low. Further, the camera stabilizer target position correction method of the present invention has a proximity target automatic tracking capability by the camera stabilizer even when the target is on the roof or intermediate floor of a building or the like and the altitude element in the target position data is missing. improves. Furthermore, the camera stabilizer target position correction method of the present invention captures the target in the camera's field of view when the camera-equipped aircraft approaches the target because the target position acquired from a map or the like is low in accuracy. Even when it is not possible to continue, the camera can automatically capture the target without depending on the vision, and thus the time from the start of the flight of the aircraft in the direction of the target to the start of imaging of the target is reduced. The time to start can be shortened regardless of the skill level of the camera operator.

  Next, an embodiment of the camera stabilizer target position correction method of the present invention will be described. FIG. 1 is a block diagram showing a camera stabilizer to which a camera stabilizer target position correcting method according to an embodiment of the present invention is applied. The camera stabilizer 10 includes a target position correction calculation unit 1, a target position holding unit 2, a tracking gimbal angle calculation unit 3, a gimbal angle control unit 4, a camera 5, an inertia device 6, and an operation unit 7. The gimbal angle control unit 4 has a gimbal mechanism. The housing of the camera 5 and the housing of the inertial device 6 are fixed to each other to form a camera / inertia device integrated structure. The camera / inertia integrated structure is mounted on the gimbal mechanism of the gimbal angle control unit 4. A target position 111 is input from the map 11 to the camera stabilizer 10, and a current position 112 is input from a GPS (Global Positioning System) 12. An operation 113 by the operator 13 who operates the camera stabilizer 10 is a manual operation input to the joystick (operation input lever) of the operation unit 7. This camera stabilizer 10 is mounted on a helicopter.

  As a means for electronically storing the three-dimensional map data in the digital format, for example, there is a three-dimensional geographic data storage device described in FIG. 11 may be digital map data stored in a three-dimensional or two-dimensional geographic data storage device having a similar configuration. Further, map information is stored as digital data in a car navigation system (a map display device called car navigation system) mounted on a car, but the map 11 is stored in the same format as that in the navigation system. The map information can be used. If the map information is digital data stored in the same format as the car navigation system, the map is displayed on the map display device similar to the map display device in the car navigation system, and the target on the map is specified with the cursor. Thus, the latitude and longitude of the target can be read out and provided as the target position 111 to the target position holding unit 2. This target position 111 represents the position of the target to be imaged by the camera 5. The map 11 may be a map drawn on paper. When the map 11 is drawn on paper, the target position 111 is the latitude, longitude, and altitude that are visually read from the map 11 by the operator 13, and is input to the target position holding unit 2 by the keyboard. This keyboard is not shown in FIG. 1 but is included in the map 11.

  The GPS 12 is mounted on a helicopter on which the camera stabilizer 10 is mounted, detects the position of the helicopter, and outputs the position of the helicopter as the current position 112 of the camera stabilizer 10. The current position 112 is made up of latitude, longitude, and altitude data of the camera stabilizer 10. Since the distance between the camera stabilizer 10 and the GPS 12 is much smaller than the distance between the camera stabilizer 10 and the target (imaging target), the position of the helicopter detected by the GPS 12 may be regarded as the position of the camera stabilizer 10. It is.

  Next, the operation of each component of the camera stabilizer 10 of FIG. 1 will be described. The target position correction calculation unit 1 inputs the pre-correction target position 102a, the current position 112 of the camera stabilizer 10, the camera visual axis direction 106, and the correction instruction 107b, and the corrected target position only when the correction instruction 107b is input. 101 is calculated and the corrected target position 101 is output to the target position holding unit 2. The calculation of the corrected target position 101 in the target position correction calculation unit 1 will be described in detail later using mathematical expressions.

  The target position holding unit 2 initially stores the target position 111 input from the map 11 when the camera stabilizer 10 is started up, and outputs the target position 111 as a pre-correction target position 102a and a target position 102b. Thereafter, when the corrected target position 101 is supplied, the target position holding unit 2 replaces the previously stored target position 111 with the corrected target position 101, and sets the corrected target position 101 as a new target position. Save and output as the pre-correction target position 102a and the target position 102b. Thereafter, each time a new corrected target position 101 is supplied, the previously stored target position is replaced with the new corrected target position 101, and the new corrected target position 101 is saved as a new target position. At the same time, they are output as the pre-correction target position 102a and the target position 102b.

  The tracking gimbal angle calculation unit 3 inputs the target position 102b, the current position 112, the camera visual axis direction 106, and the current gimbal angle 104a, calculates the tracking gimbal angle 103 based on these data, and calculates the tracking gimbal angle 103 as the gimbal. Output to the angle controller 4. The tracking gimbal angle calculation unit 3 calculates the direction in which the visual axis of the camera 5 should be directed from the target position 102b and the current position 112. Further, the tracking gimbal angle calculation unit 3 determines the difference between the visual axis direction of the camera 5 to be directed and the camera visual axis direction 106 representing the current visual axis direction of the camera 5 and the current gimbal angle representing the current gimbal angle. From 104a, a tracking gimbal angle 103 is calculated. The tracking gimbal angle 103 is an angle that each axis of the gimbal mechanism in the gimbal angle control unit 4 should take so that the visual axis of the camera 5 points to the target. The tracking gimbal angle 103 is supplied to the gimbal angle control unit 4, controls the gimbal mechanism of the gimbal angle control unit 4, performs the visual axis control 104 b by an appropriate angle, and moves the camera 5 to the target to be tracked by the camera 5. This is a signal for automatically directing the visual axis.

  The gimbal angle control unit 4 has a gimbal mechanism, performs a visual axis control 104b according to the drive command 107a or the tracking gimbal angle 103, and rotates each axis of the gimbal mechanism by the visual axis control 104b. To control. The visual axis control 104b is a mechanical output for rotating each axis of the gimbal mechanism, and controls the visual axis of the camera 5 in a direction determined by the angle of each axis of the gimbal mechanism. The gimbal angle control unit 4 performs the visual axis control 104b based on only the drive command 107a when the drive command 107a is input, and controls the visual axis control 104b based on the tracking gimbal angle 103 when there is no input of the drive command 107a. I do. The gimbal angle control unit 4 detects the angle of each axis of the gimbal mechanism as the current gimbal angle 104a, and outputs the current gimbal angle 104a to the tracking gimbal angle calculation unit 3.

  The camera 5 has a field of view within a predetermined angle range centered on the visual axis, and is a means for imaging the field of view. The inertial device 6 fixed integrally with the camera 5 detects the posture and azimuth angle of the camera 5, generates a camera visual axis direction 106 represented by this posture and azimuth angle, The camera visual axis direction 106 is output to the tracking gimbal angle calculation unit 3. The operation unit 7 is provided with a joystick as an operation input lever, and when a manual operation of the operator 13 is input with the joystick and the size of the operation exceeds a threshold value, data corresponding to the size and direction of the operation is obtained. The drive command 107 a is generated and output to the gimbal angle control unit 4. Further, when the magnitude of the operation of the operator 13 exceeds the threshold value, the operation unit 7 generates a correction instruction 107 b and outputs the correction instruction 107 b to the target position correction calculation unit 1.

  Next, the operation of the camera stabilizer 10 of FIG. 1 will be described in detail with reference to FIGS. FIG. 2 shows a camera in the camera stabilizer 10 mounted on the helicopter 100 when there is an imaging object at the rooftop position Q of the high-rise building 200 and the ground position P1 of the building 200 is the target position 111 acquired on the map 11. It is a schematic diagram showing that the visual axis of 5 is L1. In FIG. 2 (also in FIGS. 3 and 4), the position of the camera stabilizer 10 acquired from the GPS (substantially the same as the position of the helicopter 100) is N1. The camera stabilizer 10 is activated and determines the visual axis of the camera 5 based on the target position P1 on the ground surface represented by the target position 111 supplied from the map 11 and the current position 112 of the camera stabilizer 10 acquired from the GPS. Automatic tracking that automatically points to the upper target position P1 is started. When the distance from the helicopter 100 to the target position P1 is large, the position Q of the imaging object enters the field of view of the camera 5 even if the visual axis of the camera 5 is directed to the target position P1, and thus the helicopter 100 moves. However, the camera 5 can automatically capture an image of the object to be imaged. However, when the helicopter 100 is close to the target position P1, the visual axis of the camera 5 is still directed to the target position P1, so that the position Q of the imaging target deviates from the field of view of the camera 5 and the imaging target can be imaged. Disappear.

  Even in such a case, the operator 13 operates the joystick of the operation unit 7 and inputs an operation 113 having a magnitude greater than or equal to the threshold value to the operation unit 7, thereby generating the drive command 107 a and the visual axis of the camera 5. , The imaging object is captured in the field of view of the camera 5, the visual axis of the camera 5 is directed to the position Q of the imaging object, and the position Q is captured at the center of the field of view of the camera 5 (see FIG. 3). . FIG. 3 conceptually shows how the visual axis of the camera 5 is changed from L1 to L2.

  The operation unit 7 generates a correction command 107b at the same time as generating the drive command 107a. Upon receiving the correction instruction 107b, the target position correction calculation unit 1 calculates the corrected target position 101. The position represented by the corrected target position 101 is now P2 (see FIG. 4). In FIG. 4, S1 is a plane passing through the target position P1 with the visual axis L2 as a normal vector. The corrected target position P2 is an intersection of the plane S1 and the visual axis L2, and is located on the visual axis L2 and closest to the initial target position P1.

  The target position holding unit 2 changes the target position to be held from the initial target position P1 to the corrected target position P2, and the target position 102b is also set as the corrected target position P2. Then, the tracking gimbal angle calculation unit 3 and the gimbal angle control unit 4 output the visual axis control 104b so that the visual axis of the camera 5 is always directed to the corrected target position P2. Thus, thereafter, the camera stabilizer 10 automatically tracks this new target position P2.

  When the helicopter 100 moves to the position N2 in FIG. 5 while automatically tracking to the new target position P2, the visual axis is L2 ′ and the target position P2 is captured. Assume that the position Q is out of the field of view of the camera 5 again. At this time, when the operator 13 gives the operation 113 to the joystick of the operation unit 7 again and inputs an operation having a magnitude greater than or equal to the threshold value to the operation unit 7, the operation unit 7 generates a drive command 107 a and views the camera 5. By moving the axis, the imaging object is captured in the field of view of the camera 5, the visual axis of the camera 5 is directed to the position Q of the imaging object, and the position Q of the imaging object is again set at the center of the field of view of the camera 5. capture. The visual axis L3 in FIG. 6 represents the visual axis of the new camera 5 that is directed to the position Q of the imaging object. The position of the helicopter 100 in FIG. 6 is N2 as in FIG.

  Since the operation unit 7 outputs the correction instruction 107b together with the drive command 107a, the target position correction calculation unit 1 newly generates a corrected target position 101 as in the case of the output of the previous correction instruction 107b. The new corrected target position 101 at this time is set as a corrected target position P3. As shown in FIG. 6, the post-correction target position P3 is a point on the visual axis L3 and is the closest position to the previous target position P2. In FIG. 6, S2 represents a plane having the visual axis L3 as a normal line. The target position holding unit 2 stores the new post-correction target position P3 and outputs it as the pre-correction target position 102a and the target position 102b. Thereafter, the camera stabilizer 10 automatically tracks the visual axis of the camera 5 to the new target position P3 by the action of the tracking gimbal angle calculation unit 3 and the gimbal angle control unit 4 as in the previous automatic tracking of the target position P2. This new target position P3 is closer to the imaging object position Q than the previous target position P2.

  With such a procedure, in the camera stabilizer 10 of FIG. 1, the operator 13 moves the position Q of the imaging object out of the field of view of the camera 5 according to the movement of the helicopter 100, or before the operator 13 moves out of the field of view of the camera 5. By inputting the operation 113 to the operation unit 7 so as to capture the position Q of the imaging object in the center of the visual field of the camera 5 and sequentially updating the target positions (for example, P1, P2, P3 described above) for automatic tracking. The target position for automatic tracking can be gradually brought closer to the position Q of the object to be imaged.

  Next, the principle of the camera stabilizer target position correction method of the present invention will be described. For convenience of explanation, it is assumed that the camera stabilizer to which the present invention is applied has the configuration shown in FIG. FIG. 7 is a diagram for explaining the principle of the target position correction method of the present invention applied to the camera stabilizer 10 of FIG. 1, and is a conceptual diagram showing the relationship of each element in the present invention. The relationship between each element shown in FIG. 7 is that the target position that the camera stabilizer 10 has been subject to automatic tracking is the point P, and the target position P is out of the field of view of the camera 5 or is about to come off. By moving the visual axis of the camera 5 by the operation 113 by the operator 13, the imaging target is captured in the visual field of the camera 5, the visual axis of the camera 5 is directed to the position Q of the imaging target, and the imaging target Is captured at the center of the field of view of the camera 5, the visual axis L of the camera 5 is directed to the position Q of the object to be imaged, and this operation corrects the target position for automatic tracking of the camera stabilizer 10 to produce a new point. This is when T is the target position.

  In the camera stabilizer 10, even when there is a difference between the target position P1 input from the map and the actual position Q of the imaging object, correction target position calculation is performed to enable tracking of the imaging object. As described above, the calculation is performed by the target position correction calculation unit 1 and is output as the corrected target position 101. The calculation for correcting the target position for automatic tracking is performed as follows.

本発明による目標位置補正方法では、補正した後の目標位置は、視軸直線上であって、前回の目標位置に最も近い点とする。この補正後の目標位置の座標Ｔを計算するために、平面Ｓを仮想する。平面Ｓは、視軸ベクトルＶを法線とし、前回の目標位置Ｐを通る平面である。この平面Ｓは、
Ｘｖ・（ｘ−Ｘｐ）＋Ｙｖ・（ｙ−Ｙｐ）＋Ｚｖ・（ｚ−Ｚｐ）＝０・・・・・（２）
と表せる。 The coordinates of the previous target position P
P (Xp, Yp, Zp)
The coordinate N of the current position of the helicopter 100 is
N (Xn, Yn, Zn)
When the helicopter 100 is at the current position of the coordinate N and the camera 5 captures the imaging target (position Q), the vector V of the visual axis L of the camera 5 (referred to as the visual axis vector V).
V (Xv, Yv, Zv)
And At this time, the visual axis straight line representing the visual axis L of the camera 5 is expressed by the following equation (1) when an auxiliary variable t is given.

In the target position correction method according to the present invention, the corrected target position is on the visual axis straight line and is the closest point to the previous target position. In order to calculate the coordinates T of the corrected target position, the plane S is assumed. The plane S is a plane passing through the previous target position P with the visual axis vector V as a normal line. This plane S is
Xv · (x−Xp) + Yv · (y−Yp) + Zv · (z−Zp) = 0 (2)
It can be expressed.

となる。 The coordinate T of the intersection point between the visual axis straight line represented by the equation (1) and the plane S represented by the equation (2) is the coordinate of the target position after correction. Solving the simultaneous equations of Equation (1) and Equation (2) for the auxiliary variable t,

It becomes.

From this equation (3) and equation (1), the coordinate T is
(Xv.t + Xp, Yv.t + Yp, Zv.t + Zp) (4)
Is required. Thus, the corrected target position “a point on the visual axis straight line and closest to the previous target position” according to the present invention is determined by the equation (4).

  As is apparent from FIG. 7, when the distance between the position Q of the imaging object and the previous target position P is D1, and the distance between the position Q of the imaging object and the corrected target position T is D2, the position Q, Since T and P form a right triangle, D2 <D1 is always satisfied. This inequality indicates that the corrected target position T is closer to the position Q of the object to be imaged at every correction.

  As described in detail above, the embodiment of the present invention described with reference to FIG. 1 to FIG. 7 shows the visual axis L of the camera 5 when the target position P and its own position N are given. It is a position control tracking type camera stabilizer that automatically points to the target position P, the initial target position P1 is given by the map 11, and its own position N is given by the GPS 12, and the operation 113 of the operator 13 is input. Applied to a camera stabilizer in which the visual axis L is controlled in accordance with an operation 113, the visual axis L of the camera 5 is automatically directed to the initial target position P1, and the imaging target By giving the operation 113 at the first capture position N1 at which the object position Q can be captured by the camera 5, the visual axis of the camera 5 is directed to the position Q of the image capture object, and the image capture object from the first capture position N1 Turtle that points to the position Q The point closest to the first target position P1 on the five visual axes L2 is obtained by calculation as the second target position P2, and the camera visual axis L is automatically set to the second target position P2 if necessary. And by giving the operation 113 at the second capture position N2 where the position Q of the imaging object can be captured by the camera 5, the visual axis of the camera 5 is directed to the position Q of the imaging object, A point that is closest to the second target position P2 on the visual axis L3 of the camera 5 pointing from the second capture position N2 to the position Q of the object to be imaged is calculated as the third target position P3, and is necessary thereafter. Similarly, the target position is corrected so as to gradually approach the position Q of the object to be imaged by correcting the target position in the same manner.

  With this camera stabilizer target position correction method, when an aircraft equipped with a camera is close to the target, even if the accuracy of the target position acquired from a map or the like is low, the target can continue to be captured in the camera's field of view. The target position can be corrected. FIGS. 2 to 7 illustrate the case where the altitude data at the target position 111 input from the map 11 is the position P1 on the ground surface of the building 200 and not the position Q of the object to be actually imaged. . However, the present invention is always effective when the target position with the required accuracy cannot be obtained from a map or the like. For example, the present invention is effective not only when the altitude but also the accuracy of latitude and longitude data is not sufficient, and only on the rooftop of the building. In other words, the target position can be flexibly corrected when an intermediate floor is an imaging target or when the position of the imaging target changes with time.

  By adopting the camera stabilizer target position correction method of the present invention, it is possible to improve the proximity target automatic tracking capability by the camera stabilizer even when the accuracy of the target position acquired from a map or the like is low. Further, the camera stabilizer target position correction method of the present invention has a proximity target automatic tracking capability by the camera stabilizer even when the target is on the roof or intermediate floor of a building or the like and the altitude element in the target position data is missing. improves. Furthermore, the camera stabilizer target position correction method of the present invention captures the target in the camera's field of view when the camera-equipped aircraft approaches the target because the target position acquired from a map or the like is low in accuracy. Even when it is not possible to continue, even when the object to be imaged cannot be captured directly visually, the camera can automatically capture the target, and the time from when the aircraft starts to fly in the direction of the target until the target starts to be imaged can be reduced. The time until the start of imaging can be shortened regardless of the skill level of the camera operator.

  In addition, although embodiment was mentioned above and this invention was demonstrated concretely, of course, this invention is not limited to this embodiment.

It is a block diagram which shows the camera stabilizer to which one Embodiment of the camera stabilizer target position correction method of this invention is applied. 4 is a conceptual diagram showing a state in which a visual axis L1 of a camera 5 mounted on a helicopter 100 is directed to a position P1 on the ground surface of a building 200. FIG. When the visual axis L1 of the camera 5 mounted on the helicopter 100 is directed to the position P1 on the ground surface of the building 200 (FIG. 2), the operation unit (joystick) 7 of the camera stabilizer 10 is operated to view the camera 5 It is a conceptual diagram which shows a mode that the axis | shaft was changed from L1 to L2 and the visual axis of the camera 5 was orient | assigned to the position Q on the roof of the building 200. FIG. When the visual axis L1 of the camera 5 mounted on the helicopter 100 is directed to the position P1 on the ground surface of the building 200 (FIG. 2), the operation unit (joystick) 7 of the camera stabilizer 10 is operated to view the camera 5 The axis is changed from L1 to L2, the visual axis of the camera 5 is directed to the position Q on the roof of the building 200 (FIG. 3), and the target position correction calculation unit 1 of the camera stabilizer 10 determines the target position of the tracking target. It is a conceptual diagram which shows a mode that it correct | amends to the point P2. FIG. 5 is a conceptual diagram illustrating a state in which the camera stabilizer 10 directs the visual axis of the camera 5 to the corrected tracking target position P2 in FIG. 4 when the helicopter 100 moves from the position in FIG. 4 to another position. When the helicopter 100 moves from the position of FIG. 4 to another position, the camera stabilizer 10 directs the visual axis of the camera 5 to the corrected tracking target position P2 of FIG. 4 (FIG. 5), and the target position of the camera stabilizer 10 It is a conceptual diagram which shows a mode that the correction calculation part 1 correct | amends the position of the target of tracking object to the point P3. It is a figure explaining the principle of the camera stabilizer target position correction method of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Target position correction | amendment calculation part 2 Target position holding | maintenance part 3 Tracking gimbal angle calculation part 4 Gimbal angle control part 5 Camera 6 Inertial device 7 Operation part 10 Camera stabilizer 11 Map 12 GPS (Global Positioning System)
13 operator 101 post-correction target position 102a pre-correction target position 102b target position 103 tracking gimbal angle 104a current gimbal angle 104b visual axis control 106 camera visual axis direction 107a drive command 107b correction instruction 111 target position 112 current position 113 operation L visual axis Straight line N Current position of aircraft 100 P Previous target position Q Position of object to be imaged S Plane with P as normal vector V Visual axis vector T, P2, P3 Target position after correction

Claims (2)

  When a target position P and its own position N are given, it is a position control tracking type camera stabilizer that automatically points the visual axis L of the camera to the target position P, and the initial target position P1 is a map, Further, the present invention is applied to a camera stabilizer in which its own position N is given by GPS and an operation unit for inputting an operator's operation is provided, and the visual axis L is controlled in accordance with the operation. By giving the operation at the first capture position N1 that is directed to P1 and the position Q of the imaging object can be captured by the camera, the visual axis L is directed to the position Q of the imaging object, and the first A point that is closest to the initial target position P1 on the visual axis L2 of the camera that is directed from the capture position N1 to the position Q of the object to be imaged is calculated as the second target position P2, and is further viewed as necessary. The axis L is fingered to the second target position P2. And giving the operation at the second capture position N2 at which the position Q of the imaging object can be captured by the camera, the visual axis L is directed to the position Q of the imaging object, and the second capture position N2 To obtain the point closest to the second target position P2 on the visual axis L3 of the camera directed to the position Q of the object to be imaged as the third target position P3. A camera stabilizer target position correction method for correcting the target position P so as to gradually approach the position Q of the imaging object by correcting the position. This is a position control tracking type camera stabilizer that automatically directs the visual axis L of the camera to the target position P when the target position P and the current own position N are given, and inputs the operation of the operator In a camera stabilizer target position correction method that includes an operation unit and is applied to a camera stabilizer in which the visual axis L is controlled according to the operation and corrects the target position P so as to gradually approach the position Q of the object to be imaged.
The coordinates of the previous target position P
P (Xp, Yp, Zp)
The coordinates N of the current position
N (Xn, Yn, Zn)
When the camera stabilizer is at the coordinate N and the camera captures the imaging object at the position Q, the vector V of the camera visual axis L is obtained.
V (Xv, Yv, Zv)
Then, by giving the auxiliary variable t, the visual axis L is

It is expressed by the formula
The corrected target position is on the visual axis L and is the closest point to the previous target position,
Assuming that the vector V of the visual axis L is a normal line, a plane S passing through the previous target position P is assumed,
Plane S,
Xv · (x−Xp) + Yv · (y−Yp) + Zv · (z−Zp) = 0 (2)
And
The coordinate T of the intersection of the visual axis represented by the expression (1) and the plane S represented by the expression (2) is the coordinate of the target position after correction, and the simultaneous equations of the expressions (1) and (2) are For the parameter t

And the coordinate T is expressed by the equations (3) and (1).
(Xv.t + Xp, Yv.t + Yp, Zv.t + Zp) (4)
A camera stabilizer target position correction method characterized by being obtained by the following equation (4).

Images