JPH11183096A - Landing observation image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable stable determination of the landing position of a projectile by a method wherein a landing point of a flying projectile is computed to match any moment of flying time thereof when the projectile should reach a target using estimated items of trajectory and the position of the flying projectile before the landing thereof is extracted to forecast the position of the projectile at the landing moment. SOLUTION: The position of a flying projectile of an extracted image 31 in an image extraction timing signal 29 at an observation time generation processing section 42 is observed just as in a flying projectile image. From this extracted image, the flying projectile image is extracted by a flying projectile image extraction processing section 43 to determine the observation position 44 of the flying projectile image within the image. A flying trajectory items estimation processing section 45 determines a horizontal position conversion value x1 and a vertical position conversion value y1 from the observation position 44 of the flying projectile image. Using the estimated trajectory items 49 determined, a landing point forecast processing section 50 determines an estimated landing position based on a horizontal landing position conversion value xp and a vertical landing position conversion value yp. The estimated landing position thus obtained is outputted as forecast landing point 51 corresponding to the position in the image.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a landing observation system for predicting a landing position by observing a flying shell in an image in a system which requires observation of a shell shot at a target. The present invention relates to an image processing device.

[0002]

2. Description of the Related Art FIG. 11 shows an example of a system to which a landing observation image processing apparatus is applied. This system operates so that a target 2 is hit with a shell 2. A projectile 4 equipped with a firing gun 3, an imaging camera 5, and a distance measuring device 6 are mechanically and electrically linked in a positional relationship. Positioned by the interlocking mechanism 7. Viewing target 1 with imaging camera 5 in visual axis direction 8
And a distance 10 to the target 1 is measured by a telemeter 6 using a laser or the like in the same laser irradiation direction 9.
Further, the projectile 4 receives an operation command 11 so as to be directed in the direction of the target 1 captured by the imaging camera 5, and is directed by the drive control unit 12 in the direction 13 of the projectile in the direction of the target 1. The shell 2 mainly has a parabolic trajectory trajectory that is a function of the firing distance 10 and the firing speed inherent to the shell.
The trajectory calculation unit 16 calculates the azimuth angle 14 of the firing gun 3 necessary for hitting the target 1 with the projectile 10 that determines the firing distance 10 and the characteristics of the shell 2. An angle command 17 is output to the drive control unit 12. Accrual angle command 17
The firing gun 3 is directed in the direction of the accrual angle azimuth 14 by the control signal 18 from the drive control unit 12 which has received the input. When the firing trigger 19 for firing the shell 2 is input in the state set as described above, the shell 2 is fired from the firing gun 3 and flies toward the target 1. Observing where the flying cannonball 2 has landed near the target 1 is important for safety management, correction of the accrual angle for the next re-fire, etc., and the input image from the imaging camera 5 An impact observation image processing device 21 for observing the impact position using 20 is used. For the arrival observation by the arrival observation image processing device 21,
Using the input from the trajectory calculation unit 16 such as the flight time 22 required for the shell 2 obtained from the shooting distance 10 and the firing speed to reach the target 1 and the input of the firing trigger 19 giving the timing of observation, the shell 2 A method of observing the position of the shell at the time of landing from the image is used. At this time, a display image 23 obtained by synthesizing the impact position extracted from the input image 20 and the image state in the target 1 direction is output to the display unit 24.
Reference numeral 25 denotes a current angle of the firing gun 3, the firing device 4, and the imaging camera 5. Reference numeral 26 denotes launch parameters such as visual axis orientation and trajectory parameters. FIG. 12 is a block diagram showing a conventional landing observation image processing apparatus. In the figure, reference numeral 27 denotes a target tracking processing unit which receives an input image 20 from the imaging camera 5 that captures an image of the target 1, extracts the target 1 from the image, and operates to track a shooting point in the target 1. , 28 receive the input of the firing trigger 19 and output a ballistic time counter 29 which outputs an image extraction timing signal 29 after the flight time 22. 30 captures and stores the image frame at the input time of the image extraction timing signal 29 from the input image 20. An image frame memory unit that outputs an extracted image 31; 32, an impact point extraction processing unit that extracts a shell flight position at the impact time as an impact point from the extracted image 31; 33, a reference to the visual axis azimuth current angle 34 Direction and
The tracking position data 35 output from the target tracking processing unit 27 and the landing position observation data 36 from the landing point extraction processing unit 32 are displayed and synthesized on the input image 20 using symbols or the like, and the display image 23 is output. This is a display synthesis processing unit.

[0003] The conventional landing observation image processing apparatus is configured as described above and operates as follows. From the occurrence of the firing trigger 19, which is the firing time of the shell 2, the image extraction timing signal 29 after the flight time 22, which is the time at which the shell 2 reaches the target 1, is input from the landing time counter 28, so that the landing time Is taken from the input image 20 into the image frame memory unit 30. This captured image 31
Ammunition 2 is extracted from the target 1
The user can know the position of the shell at the time of landing. In addition, the display combining processing unit 33 uses the data of the extracted landing position observation data 36 and the tracking position data 35 obtained by performing tracking processing of the target 1 with the visual axis azimuth current angle 34 as a reference position, and displays the data on the input image 20. By outputting as a display image 23 synthesized and displayed with symbols or the like, it is possible to confirm the state of impact.

FIG. 13 shows a basic flight situation from the firing of the shell 2 to the impact when the effects of air resistance, wind, and the like, which become a flight disturbance, are ignored. Target 1 is target horizontal distance 37
In order to cause the shell 2 to hit the target 1, the elevation angle of the firing gun 3 is set to the anticipation angle azimuth 14 according to the firing speed of the shell 2. 38 is a shell at launch, 39 is a shell in flight, 4
0 indicates a position on the bullet flight path 41 which becomes a parabola of the bullet at the time of impact.

FIG. 14 shows a display image 23 when a shell is fired and when a shell is attached. FIG. 14A shows a display image 23 when a shell is fired, and the input image 2 is included in the display image 23.
0 indicates that the target 1 being imaged, the visual axis azimuth current angle 34, and the symbol of the tracking position data 35 are combined and displayed. In the case where the target 1 is moving or the amount of accrual correction such as bending of a shell due to wind or the like is considered, the tracking position data 35 and the visual axis azimuth current angle 34 include positions corresponding to the amount of accrual correction. Produces a deviation. FIG. 14B shows a display image 23 when a bullet is shot, and shows that the symbol of the shot position observation data 36 that is observed and extracted from the display image when the shot is fired is combined and displayed. , This gives the point of impact of the shell,
You can observe the hit point.

Further, in order to load the landing image at the time of landing into the image frame memory section 30, it is necessary to obtain the flight time 22 from the timing of the firing trigger 19 generated at the firing time of the shell to the landing of the shell. Is required. When the position in flight of the shell shown in FIG. 13 is x in the horizontal direction in the firing direction, y in the vertical direction, α is the accrual angle, v is the firing speed of shell 2, and g is the gravitational acceleration. The horizontal position x and the vertical position y at the elapsed time t from the time are obtained by the following equations.

[0007]

(Equation 1)

[0008]

(Equation 2)

At this time, assuming that the target horizontal distance 37, which is the firing distance 10, is L, and the flight time 22, which is the time at which the shell 2 arrives at the target 1, is T, L and T are respectively obtained by the following equations. Can be

[0010]

(Equation 3)

[0011]

(Equation 4)

Therefore, in the trajectory calculation unit 16, the firing speed v specific to the shell is a known parameter, and the firing range 10 is a parameter observed and input by the telemeter 6, so that the accrual angle required for a hit is obtained. α and the flight time T can be obtained.
If an image is captured in step 2, an image at the time of impact can be obtained.

[0013]

In the conventional landing observation image processing apparatus as described above, the landing point is observed by extracting an image at the landing time.

FIG. 15 shows an example of the flight of a projectile. FIG. 15A shows a case where the shell 2 hits the target 1 at the time of impact. FIG. 15B shows a case where the shell 2 has passed above the target 1 at the time of impact. FIG. 15C shows a case where the shell 2 has landed before the target 1 before the time of landing.
As described above, in the actual flight of the shell, there is a case where the shell 2 does not exist at the target 1 position at the time of impact due to the influence of the error of the set trajectory specification and the like. The shooting range 10 is the telemeter 6
The firing speed varies depending on the environmental conditions of firing of explosives, the individual difference of shells, etc., and the actual firing speed depends on the set specifications. for,
If it is smaller, it will land before the target 1 before the landing time as shown in FIG. 15C. In the conventional method of capturing the image of the landing time, the landing is observed. There was a problem that it was not possible.

FIG. 16 shows an example of an image obtained at the time of impact. FIG. 16A shows a situation where the impact light emission 98 when the shell 2 hits the target 1 is large, and in such a case, an accurate impact due to the large spread of the impact light emission is given. It is difficult to extract and observe landing points. FIG.
6 (b) shows a situation where the shell 2 is passing near the target 1. The shell 2 near the distant target 1 has a very small size and is distant, so that it is difficult to extract and observe the shot shell image 99 near the landing due to the influence of weather conditions and the like. There was such a problem.

The present invention has been made in order to solve such a problem, and even if a shell hits before the time of hitting,
Even if you are flying above the target at the time of impact, in any case, to find the position of the impact point at the impact time, extract the shell position during flight before the impact It is another object of the present invention to stably grasp the position of the shot by predicting the position of the shot at the shot time.

Further, the present invention can be applied to a case where it is difficult to observe the position of an impact due to impact light emission, and even if it is difficult to observe a distant flying cannonball on an image. To find out where the point should be, stabilize by extracting the shell position in flight before landing where the flying shell position is farthest to the target and predicting the shell position at the time of arrival The purpose of this is to make it possible to grasp the impact position.

[0018]

According to the first aspect of the present invention, a landing observation image processing apparatus is provided for a flight before a landing time, which is less affected by an actual landing state and observability deterioration due to a distant shell. Means for generating an image extraction timing signal at a certain timing, extracting the position of the projectile in flight from the image extracted by the image extraction timing signal, and extracting the position of the projectile from the position of the extracted projectile on the image of the projectile. Means are provided for estimating trajectory parameters for determining a trajectory trajectory, and for calculating a landing point at a flight time at which a flying bullet should reach a target using the estimated trajectory parameters.

Further, the landing observation image processing apparatus according to the second aspect of the present invention provides an actual landing state and n times of flight before the landing time which are less affected by observability degradation due to the distant shell. Means for generating an image extraction timing signal of
Means for taking in n frames of the extracted images in the image extraction timing signal for each time, means for extracting n flying shell positions from the extracted images, and n on the image of the extracted flying shells
Determining the trajectory of the projectile from the projectile positions
In order to estimate individual ballistics and improve observation accuracy,
Means for calculating the average value of the obtained n estimated ballistic parameters as ballistic parameters, and calculating the impact point at the time of flight at which the flying shell should reach the target using the averaged ballistic parameters Means.

Further, the landing observation image processing apparatus according to the third aspect of the present invention provides an actual landing state and a flight twice before the landing time, which are less affected by observability deterioration due to the distant shell. Means for generating an image extraction timing signal of
Means for capturing two frames of the extracted image by the image extraction timing signal of each time, means for extracting two flying shell positions from the extracted image,
The flying trajectory reaches the target using the flight trajectory straight line approximation processing means that linearly approximates the trajectory of the flying munition from the positions of the flying munitions in the bullet trajectory direction, and the linear approximation trajectory formula estimated by the flight ballistic straight line approximation processing means Means for calculating a point of impact at the time of flight to be performed.

Further, the landing observation image processing apparatus according to the fourth aspect of the present invention provides an actual landing state and a certain number of times of flight during the flight before the landing time which are less affected by observability deterioration due to the distant shell. Means for generating an image extraction timing signal of
Means for capturing m frames of the extracted images in the image extraction timing signal, means for extracting m flying shell positions from the extracted images, and m on the image of the extracted flying shells
The flying trajectory reaches the target using the flying trajectory curve approximation processing means that approximates the trajectory of the flying munition in the direction of the trajectory from the positions of the flying munitions and the curve approximation trajectory formula estimated by the flying trajectory curve approximation processing means Means for calculating a point of impact at the time of flight to be performed.

Further, the landing observation image processing apparatus according to the fifth aspect of the present invention suppresses the extraction of other clutter and improves the accuracy of the extraction of flying shells from the input image by reducing the flying shells which are moving objects in the image. As a method to extract only
A means for obtaining a difference between the previous image frame and the current image frame and taking in the difference image as an extracted image is provided.

In the hitting observation image processing apparatus according to the sixth aspect of the present invention, the range of the shot trajectory observed in the image is limited to some extent in order to improve the extraction accuracy of the shot shot from the input image. An extraction region selection means is provided for limiting the extraction processing region of the flying cannonball in the extraction image and suppressing the extraction of other clutter and the like.

In the hit observation image processing apparatus according to the seventh aspect of the invention, the size of the image observation of the flying shell is limited to some extent in order to improve the accuracy of extracting the flying shell from the input image. There is provided an extraction size selection processing means for providing an extraction size for suppressing extraction and effectively extracting only a flying cannonball, and a means for extracting a flying cannonball according to an extraction size selected from within an extraction image. .

[0025]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 FIG. 1 is a configuration diagram showing Embodiment 1 of the present invention. In the figure, 19, 2
0, 22, 23, 27, 29, 30, 31, 33, 34
And 35 are the same as the above-mentioned conventional device. An observation time generation processing unit 42 generates an image extraction timing signal 29 at a certain time during the flight time 22 from the occurrence of the firing trigger 19, and an extraction 43 extracted in the image frame memory unit 30 by the image extraction timing signal 29. A flying ammunition extraction processing unit that extracts a flying ammunition image from the image 31 and obtains and outputs a flying ammunition image observation position 44. A visual axis direction 45 given by the trajectory calculation unit 16 that becomes the visual axis direction 8 of the imaging camera 5 Using the gun specification 46 which is the parallax amount between the current angle 34 and the direction of the projectile, the horizontal and vertical positions x1, y1 of the projectile corresponding to the projectile image observation position 44 are calculated. A flight trajectory specification estimating unit for obtaining an estimated trajectory specification 49 from the x1, y1 and the known trajectory specification 47 including the observed flight time 48 and the shooting distance 10; Flying projectile using the estimated ballistic specifications 49 from the original estimation processing unit 45 is point of impact prediction processing unit for calculating a bullet deposition prediction point 51 at time to reach the target 1.

A description will be given of the operation of the above-configured landing observation image processing apparatus for calculating the landing point from the flying shell position extracted from the observation image at a certain time.
FIG. 8 is a front view of the imaging camera 5 and the firing gun 3 mounted on the launcher 4 as viewed from the target 1 direction. In general, the direction of the projectile 4 that fires the shell 2 and the direction of the visual axis of the imaging camera 5 are not coaxial, and have a parallax that is shifted as shown in the figure. FIG. 9A shows an intra-image bullet trajectory 52 of a flying bullet observed in the display image. When the configuration shown in FIG. 8 is adopted by the above-described parallax, a shell that flies in a parabolic shape in the direction of the target 1 is observed as a flying trajectory that falls in the direction of the target 1 from the upper right in the image. In addition, in the range shown by the flight locus 53 near the impact, it is observed that it falls straight down in the vertical direction. Since the parallax is constant in the applied system or the like, the relationship between the observed image position in the image and the flying position in the target 1 direction can be uniquely determined, and the horizontal position x, the vertical position, and the vertical position can be determined from the observed image position. Conversion to the directional position y is possible.

The flying bullet position of the extracted image 31 in the image extraction timing signal 29 of the observation time generation processing unit 42 is
This is observed as a flying shell image 54 shown in FIG. The flying projectile image 54 is extracted from this image by the flying projectile extraction processing unit 43, and the flying projectile image observation position 4 in the image is extracted.
Ask for 4. The flight trajectory specification estimation processing unit 45 obtains a horizontal position converted value x1 and a vertical position converted value y1 from the flying cannonball image observation position 44. Image extraction timing signal 2
Assuming that the flight time 48, which is the occurrence time of No. 9, is t1, the estimated accrual angle α 'and the estimated firing speed v' of the estimated ballistic parameters to be obtained can be obtained by the following equations.

[0028]

(Equation 5)

[0029]

(Equation 6)

Further, the estimated flight time T 'at this time can be obtained by the following equation, assuming that the original shooting distance L is the same.

[0031]

(Equation 7)

Using the estimated trajectory specifications 49 obtained as described above, the impact point prediction processing section 50 calculates the estimated impact position as a horizontal impact position converted value xp and a vertical impact position converted value yp. Each is calculated by the following equation.

[0033]

(Equation 8)

[0034]

(Equation 9)

The obtained estimated landing position is output as a predicted landing point 51 at a position in the image. In this way,
An impact observation image processing apparatus capable of determining an impact prediction point using a flying bullet image is configured.

Embodiment 2 FIG. 2 is a configuration diagram showing a second embodiment of the present invention. In the figure, 19, 20, 2
Reference numerals 2, 23, 27, 33, 34 and 35 are the same as those in the conventional apparatus. Also, 46, 47 and 51 are the same as those in FIG. 55 is an observation time n-times generation processing unit that generates an image extraction n-times timing signal 56 at a certain n times during the flight time 22 from the occurrence of the firing trigger 19,
Reference numeral 57 denotes an n-frame frame memory unit for capturing n frames of images at a time corresponding to the n-time image extraction n-times timing signal 56, and 58 an n-frame frame memory unit 5
7, n flying shell images are extracted from the extracted n images 59, and n flying shell images n observation positions 60 are extracted.
The extraction processing unit 61 for obtaining and outputting the n is a number of each of n each using the gun visual axis azimuth current angle 34 which is the visual axis azimuth 8 of the imaging camera 5 and the gun specification 46 which is the parallax amount between the flying shell ammunition directions. Conversion values x1 to the horizontal position and the vertical position in the shell projecting direction corresponding to the n flying projectile images n observation positions 60
xn, y1 to yn are obtained, and x1 to xn, y1 to yn are obtained.
And n flight trajectory specifications estimating processing for obtaining n estimated n trajectory specifications 63 from the n flight times observation time 62 corresponding to each n observations and the known trajectory data 47 including the shooting distance 10 The unit 64 uses the average value of the n estimated n trajectory specifications 63 from the n trajectory specifications estimation processing unit as the estimated trajectory data, and estimates the landing at the time when the flying shell reaches the target 1 This is an averaged landing point prediction processing unit that calculates points.

In the landing observation image processing apparatus configured as described above, the number of flying shell positions extracted from the observation image at one observation time in the first embodiment is one, whereas n The operation for obtaining the observation position of each flying shell image is the same, except that n flying shell positions are extracted at each observation time. Although the observation sampling rate of the image is the imaging frame rate of the imaging camera 5, the effect of the observation error due to the sampling time on a high-speed flying shell can be large. The purpose of using n observation data is to improve the accuracy of the impact point prediction by applying the average value to the calculation of the estimated ballistic parameters obtained from the observation values in order to reduce the influence of the error due to this sampling time. It is for planning.

The n number of n shot projectile images obtained by the n number of shot projectiles extraction processing unit 58 are defined as x1 to xn, y.
1 to yn, and n times of image extraction n times of each timing signal 5
The observation time 62, which is the n times of flight, which is the occurrence time of
To tn, the estimated accrual angle α of the estimated ballistic data to be obtained
1 ′ to αn ′, and estimated firing speeds v1 ′ to vn ′, k is 1 to
The subscripts can be obtained by the following equations.

[0039]

(Equation 10)

[0040]

[Equation 11]

The estimated n trajectory specifications 63 obtained as described above
, And the averaged impact point prediction processing unit 64 obtains the average values of the obtained n estimated accrual angles and the estimated firing speeds as α ″ and v ″, respectively. The estimated flight time T ″ at this time can be obtained by the following equation.

[0042]

(Equation 12)

Using the estimated trajectory parameters obtained as described above, the averaged impact point prediction processing unit 64 calculates the estimated impact position as the horizontal impact position converted value xp and the vertical impact position converted value yp.
Are obtained by the following equations.

[0044]

(Equation 13)

[0045]

[Equation 14]

The obtained estimated landing position is output as a predicted landing point 51 at a position in the image. In this way,
An impact observation image processing apparatus capable of highly accurately predicting an impact point by reducing the influence of an observation error caused by an observation timing using a projectile image in flight is configured.

Embodiment 3 FIG. 3 is a configuration diagram showing a third embodiment of the present invention. In the figure, 19, 20, 2
Reference numerals 2, 23, 27, 33, 34 and 35 are the same as those in the conventional apparatus. Reference numerals 46 and 51 are the same as those in FIG. 65, 70 and 72 correspond to the case where the number n of times of observation timing occurrence in the second embodiment is 2. Reference numeral 71 denotes a flying projectile image 2 which is a position of two flying projectiles
Conversion values x1, x2 and y1, y2 into the horizontal position and the vertical position in the shell flight direction corresponding to the individual observation position 70
And a flight trajectory linear approximation processing unit 73 for linearly approximating the trajectory of the flying cannonball in the bullet flight direction from the x1, x2, y1, y2 and the two flight times 72 corresponding to the two observations. Is a linear approximation impact point prediction process for calculating the impact prediction point 51 at the flight time 22 which is the time when the flying shell reaches the target 1 using the linear approximation data 74 obtained from the flight trajectory linear approximation processing unit 71 Department.

The operation of the above-described impact observation image processing apparatus will be described. As shown in FIG. 9A, as the shell approaches the target 1, the trajectory 53 near the shell landing obtained in the observation image becomes linear. FIG. 10 shows a parabolic bullet trajectory in the bullet trajectory direction and a linear approximation trajectory 75 for linearly approximating the trajectory near the impact. Further, the linear approximation trajectory trajectory 75 has unknown constants a1,
Assuming b0 and b1, the horizontal position x and the vertical position y at time t are given by the following equations, respectively.

[0049]

(Equation 15)

[0050]

(Equation 16)

Therefore, when the flight trajectory is approximated by a straight line,
Observing the two bullet firing positions at different times near the impact before the impact time during the flight time 22 from the occurrence of the firing trigger 19 to the impact, the unknown constant can be obtained. it can. If the horizontal position and the vertical position at the flight time 22 are determined from the equation of the linear approximate trajectory trajectory into which the obtained constant value is substituted, the landing prediction point 51 can be obtained.

To enable such an operation, the twice-observation-time generation processing unit 65 operates so as to generate two image extraction twice timing signals 66 near the impact time. The flight trajectory straight line approximation processing unit 71 obtains a straight line from the two flight cannonball image two observation positions 70 obtained by the two flight cannonball image extraction processing unit 68 and the flight twice observation time 72 that is the two observation times. The unknown constants a1, b0, and b1 of the trajectory equation are determined, and are output as linear approximation data 74. The linear approximate impact point prediction processing unit 73 substitutes the determined constant values a1, b0, and b1 into the equation of the linear trajectory trajectory, calculates the horizontal position and the vertical position at the flight time 22 and outputs the impact prediction point 51. To work. In this manner, an impact observation image processing apparatus capable of predicting an impact point from a linear approximate trajectory trajectory obtained by observing two points of a shell in flight, regardless of the trajectory data, is configured.

Embodiment 4 FIG. 4 is a configuration diagram showing a fourth embodiment of the present invention. In the figure, 19, 20, 2
Reference numerals 2, 23, 27, 33, 34 and 35 are the same as those in the conventional apparatus. Reference numerals 46 and 51 are the same as those in FIG. Reference numerals 76 to 81 and 83 correspond to the case where the number n of times of observation timing occurrence in the second embodiment is m. 82m is a flying bullet image m which is the position of each flying bullet
Conversion values x1 to xm and y1 to ym converted to the horizontal position and the vertical position in the shell flight direction corresponding to the individual observation position 81
And a flight trajectory curve approximation processing unit 84 for approximating the trajectory of the flying cannonball in the shell flight direction from the x1 to xm, y1 to ym, and the m observation times 83 corresponding to m observations, respectively. Is a curve approximation impact point prediction process for calculating an impact prediction point 51 at a flight time 22 that is a time at which a flying shell reaches the target 1 using the curve approximation data 85 obtained from the flight trajectory curve approximation processing unit 82 Department.

The operation of the landing observation image processing apparatus configured as described above will be described. The curved approximate trajectory trajectory of the flying cannonball is given by the following formula as a horizontal position x and a vertical position y at time t, respectively, where unknown constants are c1 and d1 to dm. The subscript added to t indicates that it is a power of t.

[0055]

[Equation 17]

[0056]

(Equation 18)

Therefore, when the flight trajectory is approximated by a curve,
An unknown constant can be obtained by observing the m-point bullet flight position during the flight time 22 from the generation of the firing trigger 19 to the impact. If the horizontal position and the vertical position at the flight time 22 are determined from the equation of the curve approximate trajectory trajectory into which the obtained constant value is substituted, the landing prediction point 51 can be obtained.

In order to enable such an operation, the observation time m-times generation processing unit 76 operates so as to generate the m-time image extraction m-times timing signal 77 during the flight of the shell. The flight trajectory curve approximation processing unit 82 is a m-th time flight processing bullet extraction processing unit
The unknown constants c1, d1 to dm of the formula of the curved trajectory trajectory are determined from the m observation positions 81 of the m flying cannonball images obtained in 9 and the m observation times 83, which is the m observation times. ,
Output as curve approximation data 85. The curve approximate impact point prediction processing unit 84 substitutes the determined constant values c1, d1 to dm into the equation of the curved trajectory trajectory, calculates the horizontal position and the vertical position at the flight time 22, and outputs the impact prediction point 51. To work. In this way, a landing observation image processing apparatus capable of predicting a landing point with high accuracy from a curve approximate trajectory trajectory obtained by observing the m point of a shell in flight, regardless of the trajectory specifications, is configured. .

Embodiment 5 FIG. FIG. 5 is a configuration diagram showing a fifth embodiment of the present invention. In the figure, components other than 86 to 91 are the same as those in the first embodiment. Reference numeral 86 denotes a previous image frame memory for storing image data one frame before the current time, and 87 denotes a relationship between the previous image frame data 88 of the previous image frame memory 86 and the input image 20 which is the image data at the current time. An image difference processing unit that calculates a difference,
Reference numeral 89 denotes a difference image frame that takes in the difference image 90 at the time according to the image extraction timing signal 29 from the observation time generation processing unit 42 and outputs the extracted difference image 91 from the difference images 90 obtained by the image difference processing unit 87. Memory.

The flying cannonball has a very small size, and it is difficult to extract only the flying cannonball from the observation image under a complicated background environment such as outdoors. If the other environment clutter is stationary, the flying cannonball is a moving object, so taking the difference between the previous image data one frame before and the image data at the current time removes the stationary environment clutter, Due to the movement, only the flying cannonball having the position change and the luminance change in the image is efficiently extracted. In this case, the two positions of the disappearance portion and the generation portion of the luminance change are emphasized. However, as the shell position at the current time, the generation portion of the luminance change becomes the flying bullet image. In the extracted difference image 91 obtained in this manner, since the flying shell part is emphasized, the flying shell extraction processing unit 43
Can accurately obtain the flying shell observation image position 44, and as a result, a landing observation image processing apparatus capable of accurately determining the landing prediction point using the flying shell image is configured. .

Embodiment 6 FIG. FIG. 6 is a configuration diagram showing a sixth embodiment of the present invention. In the figure, except for 92 and 93, it is the same as the first embodiment. Reference numeral 92 denotes an extraction area selection processing unit for selecting an extraction processing area of a flying shell, and selects only an image area where a flying shell is present from the image area of the extraction image 31 from the image frame memory unit 30, and generates a selection area extraction image 93. Output to the flying shell extraction processing unit 43.

FIG. 9B shows a flying bullet passage area 94 in the observation image, which is a region through which the flying bullets pass. As shown in the figure, the area through which the projectiles pass is limited to some extent, so when extracting projectiles from the image,
It is not necessary to extract from the entire area of the image, and by adding that the extraction processing area is narrowed down, a minute flying cannonball can be effectively extracted even in a complicated background environment. In this way, the extraction time 31 and the known trajectory data 4
7 is a selected area extraction image 93 which is a flying shell passing area 94 obtained from the type of the shell and the flight characteristics given in FIG.
Is selectively extracted and output to the flying shell ammunition extraction processing section 43, and the extraction processing is limited within the selected area extraction image 93, so that the flying shell ammunition observation image position 44 can be accurately obtained by the flying shell ammunition extraction processing section 43. As a result, an impact observation image processing apparatus is configured that can accurately determine an impact prediction point using a flying bullet image.

Embodiment 7 FIG. FIG. 7 is a configuration diagram showing a seventh embodiment of the present invention. In the figure, the components other than 95 to 97 are the same as those of the first embodiment. Reference numeral 95 denotes an extraction size selection processing unit for selectively extracting the observation size of the flying shell obtained from the flight time 48 and the type of the shell and the flight characteristics given by the known ballistic specifications 47, and 96 based on the size data of the extraction size 97 of the shell. The in-size flying ammunition extraction processing section extracts the flying ammunition corresponding to the extraction size from the extracted image 31 from the image frame memory section 30 and outputs the flying ammunition observation image position 44.

Since the observed size of a flying shell is limited to some extent by the type of shell, the flight characteristics, and the time of observation, it is possible to narrow down the extraction size in a complex background environment. It is also possible to effectively discriminate and extract flying cannonballs from other clutter. Thus, the in-size flying shell ammunition extraction processing section 96
Can accurately obtain the flying shell observation image position 44, and as a result, a landing observation image processing apparatus capable of accurately determining the landing prediction point using the flying shell image is configured. .

[0065]

According to the first aspect of the present invention, since the point of impact can be predicted from the image of the shell during flight, it is stable regardless of the state of the shell image at the time of observation of the shell, the state of the observed image of the shell, and the like. It is possible to obtain a point of impact.

According to the second aspect of the present invention, the impact point is predicted using the average value of the observation values obtained from the plurality of shell images during flight, so that the influence of the observation error due to the observation timing is reduced. It is possible to obtain a highly accurate impact point.

According to the third aspect of the present invention, since the impact point is predicted by obtaining the linear approximate trajectory trajectory from the image of the projectile during flight, the impact point can be stably obtained regardless of the trajectory data. Becomes possible.

Further, according to the fourth aspect of the present invention, since the impact point is predicted by obtaining a curve approximate trajectory trajectory from the image of the projectile during flight, the impact point can be stably obtained regardless of the trajectory data. Becomes possible.

According to the fifth aspect of the present invention, since the image of the shell in flight is emphasized by the difference image and then the extraction processing is performed, the shell can be extracted with high accuracy and a highly accurate landing point can be obtained. Becomes possible.

Further, according to the sixth aspect of the present invention, since the image of the shell in flight is narrowed down to the extraction processing area, and then the extraction processing is performed, the flying shell can be extracted with high accuracy, and the point of impact with high accuracy can be determined. It is possible to obtain.

According to the seventh aspect of the present invention, since a shell size according to a flying state and the like is designated for extraction of a shell image during flight, extraction processing is performed, so that a shell shell can be accurately extracted. It is possible to obtain a high impact point.

[Brief description of the drawings]

FIG. 1 is a diagram showing a first embodiment of an impact observation image processing apparatus according to the present invention;

FIG. 2 is a diagram showing a second embodiment of a landing observation image processing apparatus according to the present invention;

FIG. 3 is a diagram showing Embodiment 3 of an impact observation image processing apparatus according to the present invention.

FIG. 4 is a diagram showing a fourth embodiment of a landing observation image processing apparatus according to the present invention;

FIG. 5 is a diagram showing a fifth embodiment of a landing observation image processing apparatus according to the present invention;

FIG. 6 is a diagram showing a sixth embodiment of the landing observation image processing apparatus according to the present invention;

FIG. 7 is a diagram showing a seventh embodiment of a landing observation image processing apparatus according to the present invention;

FIG. 8 is a front view of the launcher according to the present invention.

FIG. 9 is a diagram showing a projectile trajectory in an image of a flying projectile according to the present invention.

FIG. 10 is a diagram showing a linear approximation trajectory of a flying shell trajectory according to the present invention.

FIG. 11 is a diagram illustrating an example of a system to which the hit observation image processing apparatus is applied.

FIG. 12 is a diagram showing a conventional landing observation image processing apparatus.

FIG. 13 is a diagram showing a basic flight situation from firing of a shell to landing.

FIG. 14 is a view showing a display image in a conventional shell observation.

FIG. 15 is a diagram showing various flight examples of a shell.

FIG. 16 is a diagram showing a display image at the time of hitting in conventional artillery observation.

[Explanation of symbols]

1 target, 2 shells, 3 firing guns, 4 launchers, 5 imaging cameras, 6 telescopes, 7 interlocking mechanisms, 8 visual axis orientations,
9 Laser irradiation direction, 10 shooting distance, 11 operation commands,
Reference Signs List 12 drive control unit, 13 launcher pointing direction, 14 prevailing angle azimuth, 15 projectile type, 16 trajectory calculation unit, 17 prevailing angle command, 18 control signal, 19 firing trigger, 20
Input image, 21 shot observation image processing device, 22 flight time, 23 display image, 24 display unit, 25 current angle, 2
6 launch specifications, 27 target tracking processing unit, 28 landing time counter, 29 image extraction timing signal, 30 image frame memory unit, 31 extracted image, 32 landing point extraction processing unit, 33 display synthesis processing unit, 34 visual axis Azimuth current angle, 35 tracking position data, 36 shot position observation data, 37 target horizontal distance, 38 shot shell, 39 flying shell, 40 shot time shell, 41 shell flight path,
42 Observation time generation processing unit, 43 Flight ammunition extraction processing unit, 44 Flight ammunition image observation position, 45 Flight trajectory specification estimation processing unit, 46 Gun specification, 47 Known trajectory specification, 48
Flight time, 49 Estimated trajectory data, 50 shot point prediction processing unit, 51 shot point, 52 shot trajectory in image
3 Flight trajectory near the shell landing, 54 Flight shell image, 55
Observation time n-time generation processing unit, 56 image extraction n-time timing signal, 57 image n-frame memory unit, 58 n-projectile ammunition extraction-processing unit, 59 extracted n-image, 60 n-projectile image n observation position, 61 flight Projection unit for n projectiles, 62 observation times for n flight, 63 estimation for trajectory for n, 64 averaged impact prediction processing unit, 65 generation processing unit for twice observation time, 66 timing signal for image extraction twice, 67
2 images frame memory unit, 68 flying cannonball 2 extraction processing unit, 69 extracted n images, 70 flying cannonball image 2 observation positions, 71 flying cannonball straight line approximation processing unit, 72 flying2
Observation time, 73 straight line impact point prediction processing unit, 74
Linear approximation specifications, 75 Linear approximation shell trajectory, 76 observation time m generation processing unit, 77 image extraction m times timing signal, 78 image m frame memory unit, 79 flying m shells m extraction processing unit, 80 extracted m images 81, observation position of m flying shell images m, 82 flying shell curve approximation processing unit, 8
3 Observation time m times, 84 curve approximation impact point prediction processing unit, 85 curve approximation specifications, 86 previous image frame memory, 87 image difference processing unit, 88 previous image frame data, 89 difference image frame memory, 90 difference image , 91 extraction difference image, 92 extraction region selection processing unit,
93 Selected area extraction image, 94 Flight cannonball passage area, 9
5 Extraction size selection processing unit, 96-size projectile projectile processing unit, 97 extraction size, 98-projection emission, 99-projection projectile image.

Claims (7)

[Claims] 1. An input image from an imaging camera is input,
A tracking processing unit for directing the shooting target azimuth, an observation time generation processing unit for generating an observation time during flight of the flying object from the launch time of the flying object once, and 1 output from the observation time generation processing unit An image frame memory unit that captures an image of the observation time timing, a flying shell ammunition extraction processing unit that extracts a flying shell position from observation image data on the image frame memory unit, and a flight extracted by the flying shell ammunition extraction processing unit Using a flight trajectory specification estimating processing unit for obtaining a trajectory specification for determining a trajectory trajectory of the flying cannonball from the position of the flying cannonball on the image of the cannonball, Point prediction processing unit that calculates the point of flight at the time of flight when the flying cannonball hits the target, and the tracking information output from the tracking processing unit and the landing point prediction information from the point-of-arrival prediction processing unit And imaging turtle Bullet deposition observed image processing apparatus characterized by comprising a display synthesis processing unit for displaying output to the symbol combiner display unit on the input image of the visual axis orientation information from the imaging camera. 2. An observation time n-times generation processing unit that generates n times of observation of a flying object during flight from a launch time of the flying object, and n observation times output from the n times of observation time generation processing unit An n-frame frame memory unit for capturing the timing image, an n-shell projectile extracting unit for extracting n-shell projectile positions from the observation image data in the n-frame frame memory unit, and the n-shell extraction process A processing unit for estimating n trajectory parameters for determining a trajectory parameter for determining a trajectory of the trajectory from the n trajectory positions on the image of the trajectory extracted by the processing unit; An averaged ballistic point prediction processing unit that calculates a ballistic point at the time of flight when the flying shell lands on the target using the average ballistic parameters of the ballistic parameters estimated by the n-element estimation processing unit. 2. The impact observation according to claim 1, wherein Image processing device. 3. An observation time twice generation processing unit for generating twice the observation time during flight of the flying object from the launch time of the flying object, and two observation times output from the observation time twice generation processing unit A two-image frame memory unit for capturing an image at a timing, a two-projectile projectile extraction processing unit for extracting two flying projectile positions from observation image data on the two-image frame memory unit, and the two-projection projectile extraction The flight trajectory straight line approximation processing unit that linearly approximates the trajectory trajectory of the flying cannonball from the position of each of the two flight cannonballs on the image of the flying cannonball extracted in the processing unit, 2. The bullet according to claim 1, further comprising: a linear approximation impact point prediction processing unit that calculates an impact point at a flight time at which the flying ammunition lands on the target using the estimated linear approximation trajectory formula. Landing observation image processing device. 4. An observation time m-times generation processing unit for generating an observation time during flight of a flying object from an emission time of the flying object m times, and m observation times output from the observation time m-times generation processing unit An m image frame memory unit for capturing an image of timing, an m flying projectile extraction processing unit for extracting m flying projectile positions from observation image data in the m image frame memory unit, and the m flying projectile extraction The flight trajectory curve approximation processing unit that approximates the trajectory trajectory of the flight cannonball in the shell flight direction from the m positions of the mmunition projectiles on the image of the flight cannonball extracted by the processing unit, and the flight trajectory curve approximation processing unit 2. The bullet according to claim 1, further comprising a curve approximated point prediction processing unit that calculates a point of impact at a flight time at which the flying bullet lands on the target using the estimated curve approximated ballistic formula. Landing observation image processing device. 5. A process for capturing a flying cannonball image, wherein a difference between a previous image frame memory for storing image data one frame before the current time and image data at the current time and the previous image frame is stored. 2. The landing observation image processing according to claim 1, further comprising: an image difference processing unit for calculating, and a difference image frame memory for taking in the image difference image data at an observation timing from the observation time generation processing unit. apparatus. 6. An impact observation apparatus according to claim 1, further comprising an extraction area selection processing section for restricting an area for performing a process of extracting a flying shell from the observation image data in the image frame memory section. Image processing device. 7. An extraction size selection processing unit for restricting the size of bright spot extraction on an image for the extraction processing of a flying cannonball,
2. An impact shooting ammunition according to claim 1, further comprising an in-size flying ammunition extraction processing section for extracting a flying ammunition corresponding to an extraction size limit from the extraction size selection processing section from an image to obtain a flying ammunition position. Observation image processing device.