JPH0574025B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an automatic tracking device for tracking a moving object.

[Prior art]

BACKGROUND ART Conventionally, a moving object such as an aircraft that moves at high speed has been automatically tracked and its azimuth angle and movement angular velocity detected and used as information for controlling various devices.

[Problems to be solved by this invention]

Under these circumstances, devices that generate data regarding a moving object as a target object from a video signal including the target object obtained by an imaging device and perform automatic tracking are also used.

However, in conventional devices, processing is performed every sampling time (for example, 1/30 second in NTSC) to create one image using scanning lines, but fixed binarization is used as an image processing method. Because it separates the background and moving objects by using a digital camera, it is difficult to respond adaptively to changing environmental conditions, making it impossible to use it outdoors.

In addition, although there are models equipped with algorithms that adaptively respond to changes in environmental conditions and separate the background and moving objects, they do not allow for the separation of each piece of information that makes up the screen (for example, 250 x 250 pixels). In this case, approximately 60,000 pieces of data are processed by software, which takes several seconds to several tens of seconds per screen, making it unsuitable for measuring moving objects.

Furthermore, because the imaging device in conventional devices is fixed, it is impossible to measure moving objects if they deviate from the angle of view of the imaging device, making it impossible to track moving objects that move at high speed. .

In addition, with conventional processing equipment, if a moving object passes behind an obstacle such as a chimney, it may become impossible to identify the moving object in front or behind it, and the tracking function may stop or malfunction. equipment had to be adjusted or restarted.

Therefore, the present invention uses an algorithm that can adapt to environmental changes and reduce the number of data processed by software to several hundred pieces per screen by converting part of the processing into hardware. By improving the disadvantages of conventional devices such as Even if tracking becomes impossible due to obstructed visibility, the situation before and after that point is judged, and images are taken in the predicted direction based on the target direction and angular velocity data from one sampling time before the point in time until tracking can be restored. An object of the present invention is to provide an automatic tracking device for a moving object that can control the device and continuously perform tracking operations without restarting or intervening by an operator.

[Means to solve the problem]

The automatic tracking device for a moving object according to the present invention includes an imaging device that images a moving object, a drive stand that supports the imaging device freely in the imaging direction, and a video signal inputted from the imaging device and generated by the video signal. The moving object and the background are separated for each screen, and geometric center of gravity data indicating the position of the geometric center of gravity on the screen and the angle between the direction of the geometric center of gravity of the moving object and the center of the screen and the inside of the screen of the moving object are calculated. An image processing device that calculates the size of a moving object is input, and the size, geometric center of gravity data output from this image processing device, and azimuth data indicating the imaging direction that are input separately are input to calculate the target azimuth calculation value of the moving object. an azimuth calculation device that outputs a target azimuth calculation value, and outputs a control signal to the driving base to control the imaging direction so that the moving object is located at the center of the screen of the imaging device by inputting the target azimuth calculation value; In the tracking device, the tracking device includes a drive platform control device that outputs angle data to the position calculation device, and the azimuth calculation device further calculates the movement calculated at each sampling time by the image processing device in parallel with the operation of the tracking device. The rate of change in the size of the moving object within the screen and the amount of change in the position of the geometric center of gravity are calculated from the size of the object and the position of the geometric center of gravity, and the rate of change and amount of change exceed a predetermined tracking range setting value. Sometimes, it is determined that tracking is impossible, and the drive platform control device predicts the target direction based on the target direction one sampling time ago and angular velocity data obtained by time differentiation of the rotation angle of the rotation angle of the support shaft of the drive platform that drives the imaging device. The value is output, and the size of the moving object and the position of the geometric center of gravity are determined at each sampling time from that point onward, and the position and size of the geometric center of gravity immediately before the tracking abnormality occurs, as well as the elapsed time after the tracking abnormality occurs. The tracking device has a function of comparing the recovery range calculated from , and determining that recovery is possible only when the recovery range is within the set value of the recovery range, and operating the tracking device.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be explained below with reference to illustrated embodiments. FIG. 1 shows a block diagram of an apparatus configuration of an embodiment of an automatic tracking apparatus for a moving object according to the present invention.

In this embodiment, a TV camera 1 is used as an imaging device.
is supported on a support shaft (not shown) of the camera drive stand 2 so that it can rotate horizontally and tilt down vertically, making it possible to freely adjust the imaging direction.

Here, the TV camera 1 is supported by the camera drive base 2 so that horizontal rotation is in the X direction on the screen, and the downward direction is in the Y direction on the screen.

The video signal 3 obtained from the TV camera 1 is input to the image processing device 5 provided at the next stage, and the geometric center of gravity data of the moving object in the TV video (X and Y directions from the center of the screen to the geometric center of gravity) is input to the image processing device 5 provided at the next stage. The angle, the number of pixels indicating the length in the X and Y directions indicating the position of the geometric center of gravity from the center of the screen), and the size 9 are calculated.

Furthermore, the TV camera 1 has a built-in zoom mechanism that automatically zooms the image of the moving object 16 to the degree necessary for digital image processing within the image processing device 5, and outputs the angle of view signal θ at each sampling time. Output.

Here, the angle of view refers to the field of view that can be captured by an imaging device, and is uniquely determined from the size of the CCD used in the imaging device and the focal length of the lens.

Also, this angle of view is expressed as both horizontal and vertical angles, but since the angles in these two directions are treated in the same way, the following explanation will only explain the angle of view in the horizontal direction, but the actual angle of view is In terms of processing, both horizontal and vertical angles of view are processed simultaneously.

Normally, when a zoom lens is used, the focal length f (mm) on the zoom lens can be easily measured based on the physical positional relationship of the lens. Once this f value is determined, the angle of view can be calculated as θ=2tan ^-1 d/2f (where dmm is the horizontal length of the CCD). In this way, a signal representing f is actually input, but theoretically θ calculated from this f is used, so the following description was made as the angle of view signal θ.

By using this angle-of-view signal θ, the image processing device 5 can calculate the direction of the geometric center of gravity of the moving object by converting it into an angle from the center of the screen.

For example, if the horizontal pixels (X direction) of the screen are 256 pixels, and the position of the geometric center of gravity of a moving object is shifted from the center of the screen by 20 pixels in the X direction, the angle of view of the imaging device at this time is 10 pixels. °, the value of the geometric center of gravity data is 20 pixels, and 20 (X direction deviation pixels) / 256 (X direction screen pixels) ×
It is expressed as 10° (angle of view) = 0.78°.

Normal TV image signals (RS170 standard) use a 2:1 interlace system, so 1/30
Data for one screen can be obtained in seconds, but if the vertical resolution is set to 250 pixels or less, data can be obtained in 1/60 of a second.

Here, the image processing device 5 will be explained based on the detailed block diagram shown in FIG.
is input into the lookup table 10, and is first binarized into a moving object and a background in the first 1/60 seconds.
It is output to the XY axis projection calculator 11. The XY-axis projection calculator 11 outputs XY-axis projection data of the moving object to the image calculation device 13 from the binarized input data. The image processing unit 13 calculates the median value of the XY axis projection data in the next 1/60 seconds, uses it as the position of the geometric center of gravity, and further calculates the screen from the angle between the geometric center of gravity and the center of the screen and the position of the geometric center of gravity. Outputs the number of pixels to the center as geometric centroid data. Calculation of the geometric center of gravity data and the size 9 within the image processing device 5 is performed every 1/30 seconds, which is one sampling time.

Further, in the image processing device 5, the video signal 3 is input to the brightness histogram calculator 12 in parallel with the calculation of the geometric center of gravity data 9, and the first 1/
A brightness histogram 17 is calculated for 60 seconds and output to the image calculation device 13. The image calculation device 13 calculates lookup table data in the next 1/60 second based on the input luminance histogram 17, and updates the data in the lookup table data 10.

To explain in detail the method for determining lookup table data, a background area 15 representing the background around the moving object 16 is provided in the video screen 14 as shown in FIG. An inner luminance histogram 17 is obtained. This calculation is performed by the brightness histogram calculator 12 described above. In the histogram calculated in this way, as shown in Fig. 4b, for luminance values that are greater than or equal to a certain threshold value 19, the luminance values are 20 representing background pixels, and for luminance values that are less than or equal to the threshold value, the luminance values are representative of moving object pixels. Lookup table data is generated so that 21 correspond to each other.

In the image processing device 5 shown in FIG. 2, the lookup table 10, the XY-axis projection calculator 11, and the brightness histogram calculator 12 are configured by hardware, so the XY projection data and the brightness histogram 17 are calculated in real time. be done. Therefore,
For example, in the case of image information with 250 x 250 pixels and 64 brightness levels, the number of data to be processed by the image processing device 13 operated by software is X-axis projection data, Y-axis projection data,
The number of axial projection data and brightness histogram 17 is 250, 250, and 64, respectively, allowing high-speed processing that was impossible with the conventional method of processing each pixel data (approximately 60,000 pieces in this example) using software. Realized.

In this way, the geometric center of gravity data (angle, position) and size 9 calculated by the image processing device 5 are output to the azimuth calculation device 6 provided at the next stage. The azimuth calculation device 6 calculates a target azimuth calculation value, which is the azimuth of the target object, from the geometric center of gravity data 9 and the azimuth data indicating the imaging azimuth of the TV camera, and outputs it to the drive platform control device 7. The drive stand control device 7 controls the camera drive stand 2 that supports the TV camera 1, and controls the camera drive stand 2 so that the imaging direction of the TV camera 1 faces the target azimuth calculation value from the azimuth calculation device 6. A drive table control signal 4 is output to the drive table control signal 4. Thereby, the azimuth of the TV camera 1 is controlled so that the geometric center of gravity of the moving object is located at the center of the screen.

In addition, the direction calculation device 6 is connected to the driving platform control device 7.
Azimuth data indicating the current imaging direction of the TV camera 1 is output. Therefore, the azimuth calculation device 6 calculates the target azimuth of the moving object from the azimuth data of the TV camera 1 and the geometric center of gravity data 9 on the video screen.
Measurement data 8 such as calculated values is output.

Here, in order to prevent tracking failure or malfunction when a moving object enters behind a chimney, etc.,
The azimuth calculation device 6 is provided with a function to detect failure of automatic tracking and prevent malfunction of the device, and a function to determine the possibility of automatic tracking recovery and restore normal tracking.

Among these, the detection that automatic tracking is no longer possible is performed using the calculated value K _p of the permissible size change rate within the screen of the moving object, which is preset in the azimuth calculation device 6.
This is done by comparing the calculated value ΔX _p of the allowable positional change amount with the actual measured value.

The size of the moving object within the screen referred to here refers to the length (number of pixels) of the moving object projected on the X-axis and Y-axis within the screen, and in the following explanation, the projected length of the X-axis is used as an example. I will explain about it.

That is, when a moving object is automatically tracked normally, changes in the size and position of the moving object within the imaging screen are calculated based on its moving speed, maximum motion acceleration, and mechanical zooming speed of the imaging device 1. Since the upper limit value is limited by the value, the upper limit value is set in advance as an allowable value, and if the actual measured value exceeds the allowable value, it is determined that automatic tracking is impossible.

For example, the size of the object in the screen at a distance l ₁
Wl ₁ , the size of the object in the screen at distance l ₂ Wl ₂ ,
If the average moving speed of the object between the two points is v, and the sampling time of the imaging screen is t _s , the rate of change in the size of the object between two consecutive screens (one sampling time) _{is K 1 =} (Wl ₁ /Wl ₂ ) ^v

Claims (1)

[Scope of Claims] 1. An imaging device 1 that images a moving object 16 and outputs a video signal 3; The imaging device 1 supports the imaging device 1 freely in the imaging direction, and controls the imaging device 1 based on a control signal 4 inputted from the outside. a driving table 2 that controls the imaging direction of the image pickup device 1; and a driving table 2 that inputs the video signal 3 from the imaging device 1, separates the moving object 16 from the background 15 every sampling time of the video signal 3, and determines the geometry of the moving object 16. an image processing device 5 that calculates geometric center of gravity data consisting of the position of the academic center of gravity and the angle between the geometric center of gravity and the direction of the center of the screen and the size of the moving object in the screen; and an azimuth that indicates the imaging direction of the imaging device 1. an azimuth calculation device 6 that inputs the angle data, the geometric center of gravity data output from the image processing device 5, and the size of the moving object, and outputs a target azimuth calculation value indicating the azimuth of the moving object; A control signal 4 for controlling the imaging direction so that the moving object is located at the center of the screen of the imaging device 1 is outputted to the driving base 2 by inputting the target orientation calculation value of the moving object from the imaging device 1. In the tracking device for a moving object, the driving platform control device 7 outputs azimuth data indicating the imaging direction of the azimuth to the azimuth calculation device 6, the azimuth calculation device 6 outputs the azimuth data indicating the imaging direction of the image processing device 5 at every sampling time. Calculated geometric center of gravity data, size, and drive platform control device 7 of the moving object
a change amount calculating means for inputting azimuth data from the moving object to calculate the change rate of the size of the moving object within the screen and the change amount of the position of the geometric center of gravity; Input the calculated value of the change rate and the amount of change in the position of the geometric center of gravity, and if the rate of change in the size and the amount of change in the position of the geometric center of gravity are within the predetermined tracking range setting value, a normal tracking signal is output, and A tracking abnormality determining means that determines a tracking abnormality when the set value is exceeded and outputs a tracking abnormality signal, and a tracking abnormality signal from the tracking abnormality determining means is input to detect the movement one sampling time before the tracking abnormality occurs. Using the target azimuth calculation value of the object and the angular velocity obtained by time differentiation of the rotation angle of the drive platform support shaft that drives the imaging device calculated from the above azimuth data, the predicted target azimuth value of the moving object is calculated based on the constant angular velocity prediction. a predicted value output means for calculating and outputting this value to the drive platform control device 7 as the target azimuth calculation value; and after the operation of the predicted value output means, the geometric center of gravity data of the moving object and the size are Based on the position and size of the geometric center of gravity of the moving object one sampling time before the tracking abnormality, the size and size of the moving object are calculated at each sampling time that has elapsed since the tracking abnormality occurred. Calculate the permissible range of the position, compare the size of the moving object and the position of the geometric center of gravity with the permissible range from the reading means, determine that tracking can be restored only when the permissible range is within the permissible range, and send a recovery signal. a recovery determining means to output, and when the tracking abnormality determining means outputs a normal tracking signal and when the recovery determining means outputs a recovery signal,
The target orientation of the moving object is calculated from the azimuth angle data indicating the imaging direction of the imaging device 1 input at that time and the geometric center of gravity data output from the image processing device 5, and the calculated target orientation calculation value is calculated. 1. An automatic tracking device for a moving object, comprising: a target azimuth calculation means for outputting a target direction to a drive platform control device.