JPH0310912B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a signal processing device for a video radar that converts the brightness of a radar image acquired by a video radar.

[Conventional technology]

First, the operation of the video radar will be explained with reference to FIG. In the figure, 1 is a video radar, 2
is an antenna, 3 is a transmitter, 4 is a receiver, 5 is a signal processing device, 6 is a display device, 7 is an antenna beam,
8 is a video object, 9a and 9b are pixels (Piccell), 10 is a transmitted radio wave, and 11 is a received radio wave. The video radar 1 is generally operated in either real aperture mode or synthetic aperture mode, but the functions of the video radar signal processing device according to the present invention are completely common to both modes, and in the following description, For the sake of simplicity, the above-mentioned video radar 1 will be explained as being operated in real aperture mode.

The video radar 1 radiates transmitted radio waves 10 generated by a transmitter 3 toward a video target 8 via an antenna 2. Normally, the antenna beam 7 of the video radar 1 has extremely high directivity, and the transmitted radio wave 10 hits only a portion of the video target 8. The antenna footprint created by the antenna beam 7 on the image target 8 corresponds to one pixel of the radar image displayed on the display device 6, and is called a pixel. The transmitted radio waves 10 are scattered on the surface of the image object 8, generating scattered waves in all directions. Among them, radio waves scattered in the direction of video radar 1 are received radio waves 11.
The signal is again captured by the antenna 2 and input to the receiver 4. The receiver 4 detects the received radio wave 11, extracts the received power P which is the intensity of the received radio wave 11, and transfers this to the signal processing device 5.
The signal processing device 5 converts the above-mentioned received power P into numerical data and stores this on its internal storage means. The magnitude of the received power P stored here corresponds to the brightness of the radar image displayed on the display device 6. Hereinafter, the above received power P will be referred to as video brightness. Then, the image brightness is stored in the display device 6. Note that the address on the storage means above corresponds to the directional direction of the antenna beam 7.

The video radar 1 repeatedly performs the above operations while scanning the antenna beam 7 over the video target 8 in a two-dimensional manner. However, the antenna beam 7 scans each pixel, for example, pixel 9a and pixel 9.
This is done so that ``b'' and ``b'' do not overlap with each other. As a result, pixels are arranged two-dimensionally on the image object 8 as shown in FIG. 4, and the image brightness corresponding to each pixel is stored in the storage means as numerical data. Since the stored address corresponds to the directional direction of the antenna beam 7, the geometric relationship of each image brightness matches that of each pixel.

The radar image is obtained by transferring the numerical data stored in the storage means described above to the display device 6, converting it into a video signal, and displaying it on the screen. However, the range of numerical values that the image brightness can take is extremely large, hundreds of times the range of brightness that can be displayed by a general display device, so the signal processing device 5 converts the image brightness by some method, After the range of converted values falls within the display range of the display device, it is necessary to transfer the image brightness to the display device 6.

Hereinafter, a video luminance conversion method using a conventional signal processing device will be explained with reference to FIG. FIG. 5 shows the configuration of a conventional signal processing device 5 and a display device 6.
In the figure, 12 is an image memory which is a storage means for image brightness, and 13a to 13 are shown in FIG.
d is a register, 14 is a divider, 15 is a clipper,
16 is a multiplier, 17 is a video memory, 18 is a scan converter, 19 is a TV monitor, and 20 is an observer.

First, the numerical values P of the image brightness stored in the image memory 12 are sequentially transferred to the register 13a according to a preset order from the image memory 12. On the other hand, the TV monitor 19 is stored in the register 13b.
generated based on the numerical value α set externally by the observer 20 who views the radar image displayed on the
Pmax/α is stored. Here, Pmax is a numerical value for displaying the video brightness at maximum brightness on the TV monitor 19, and is a second constant shown in the claims. Numerical value Pmax/α stored in register 13b
is set externally by the observer 20 and stored in the register 13.
It is determined by the quotient of the value α stored in d and the value Pmax that causes the TV monitor 19 to shine at maximum brightness. The quotient of α and Pmax is calculated by the divider 14. The video luminance P stored in the register 13a and the luminance conversion constant stored in the register 13b
Pmax/α is input to the multiplier 16, and the multiplication result value P·Pmax/α is transferred to the clipper 15. The clipper 15 described above has a function expressed by the following equation (1).

y=x, OxPmax Pmax, x>Pmax (1) x: Input of the clipper y: Output of the clipper Pmax: Numerical value that makes the TV monitor 19 shine at maximum brightness In other words, the clipper 15 receives the input P.
If the value of Pmax/α is smaller than the value Pmax that makes the TV monitor 19 shine at its maximum brightness, P・Pmax/α
The value of is output, and on the other hand, the input P・Pmax/α
If the value of is greater than Pmax, the value of Pmax is output. The output y of the above clipper 15 is transferred to the video memory 17 in the display device 6 and stored therein. The address on the video memory 17 where the output y is stored is the address when the video luminance P was read out from the image memory 12. Next, the above output y is transferred from the video memory 17 to the scan converter 18, and the scan converter 18
After being converted into a video signal by the D/A converter function, the signal is displayed on the TV monitor 19 as an image. At this time, the maximum brightness on the TV monitor 19
Pixels displayed at Pmax have an image brightness of P
satisfies the following equation.

Pα (2) As is clear from equation (2), if α is made smaller, (2)
The number of pixels that satisfy the formula increases, and the overall radar image appears visually brighter. On the other hand, when α is increased, the opposite occurs, and the entire radar image appears visually dark. In many cases where the observer 20 uses radar images, the observer 20 uses radar images to detect a relatively bright isolated target from a relatively dark background in the image, and in such a usage method, the value of α is 20 greatly influences the probability of detecting an isolated target. For example, if the value of α is made too small, the image brightness P corresponding to most pixels will satisfy equation (2), and as a result, most pixels of the image object 8 will be displayed with the same maximum brightness. , observer 20
The probability that the target will be able to detect an isolated target is significantly reduced. In order to maintain a high probability of detecting isolated targets,
It is necessary to always set the value of α optimally, and in conventional signal processing devices, as shown in FIG.
The observer 20 pays attention to the radar image on the TV monitor 19, and the observer 20
The value of α was manually set based on the judgment of

[Problem that the invention seeks to solve]

In the conventional video radar signal processing device as described above, the optimal brightness conversion for detecting isolated targets is left to the radar image observer's hands, so it is not possible to provide the observer with a radar image that has been optimally brightened. In order to obtain the optimal radar image, it is necessary to provide the observer with a non-optimized radar image at least once. The problem was that it took too much time.

This invention was made to solve this problem, and is capable of providing radar images that maximize the probability of an isolated target being detected by an observer automatically and in a short period of time without the observer's intervention. The purpose is to obtain a radar signal processing device.

[Means for solving problems]

The video radar signal processing device according to the present invention includes:
maximum value detection means for detecting a pixel having the maximum value among the video luminances stored in the storage means; average calculation means for calculating the average value of the video luminances of pixels in the vicinity of the detected pixel; A multiplication means for calculating the product of the average value and a preset coefficient is provided.

[Effect]

In this invention, the signal processing device automatically finds an optimal coefficient for compressing the range of possible values of the image brightness from the image brightness stored in the storage means, so that the observer does not have to bother. It is possible to display a radar image that maximizes the probability that an observer can detect an isolated target without any problems and in a short time.

〔Example〕

FIG. 1 is a block diagram showing one embodiment of the present invention, and 6, 12 to 20 are completely the same as the conventional device described above. 21 is a switch for switching the transfer destination of the image brightness read out from the image memory 12;
22 is a maximum value detector which is a means for detecting the maximum value of image brightness; 23 is an average value calculator which is a means for calculating the average value of the brightness of pixels in the vicinity of the pixel having the maximum brightness; 24 is a register; 25 is a multiplier that is a means for multiplying the calculation result of the average value calculator 23 by the coefficient K stored in the register 24; 26 is a controller that controls the read address of the image memory 12; Note that the above coefficient K is the first constant shown in the claims.

In the video radar signal processing device configured as described above, first, the contact point A of the switch 21 is
and contact O are connected, and a series of video luminances read from the image memory 12 are transferred to the maximum value detector 22. The maximum value detector 22 detects the internally stored video brightness P _I and the transferred video brightness P _T
Compare the magnitude of and calculate P _I only if P _I < P _T.
The image brightness stored internally is converted by replacing it with _PT , and the address on the image memory 12 where _PT was stored is stored. As a result, after all the image brightnesses in the image memory 12 are transferred to the maximum value detector 22, the address of the pixel having the maximum brightness is detected by the maximum value detector 22. The detected address is the controller 26
will be forwarded to.

Next, controller 26 reads from image memory 12 only the image brightness of pixels in the vicinity of the pixel with maximum brightness. Now, if the address of a pixel is expressed as (i, j) in the form of position coordinates, and the image brightness corresponding to that pixel is expressed as Pij, then
The image brightness read out from the image memory 12 according to instructions from the controller 26 is given by the following equation.

Z=Pij I-niI+n J-njJ+n (3) In equation (3), (I, J) is the address of the pixel with the maximum brightness, and n is a constant that determines the range of the neighborhood of the pixel with the maximum brightness. . Before the signal processing device 5 starts operating, n is set to an appropriate value in the controller 26 depending on the size of the isolated target that the observer 20 wants to detect.

When the image brightness Pij given by the above equation (3) is read out from the image memory 12, the switch 2
1, contact O and contact B are connected, and at this time, the image luminance Pij is transferred to the average value calculator 23.
The average value calculator 23 calculates the input sum (2n+
1) Calculate the average brightness Ap by calculating the following formula for ^{the two} video brightness Pij, and then calculate the average brightness Ap.
is transferred to the multiplier 25.

Ap=1/(2n+1) ² _I+o 〓 ^i=In _J+o 〓 ^j=Jn Pij (4) The multiplier 25 calculates the above average luminance that has been transferred.
The multiplication of Ap by the coefficient K stored in the register 24 is performed, and the calculation result α=KAp is transferred to the register 13d and stored. The above α stored in the register 13d is a basic constant for converting the image brightness stored in the image memory 12, and in conventional signal processing devices, it is set externally by the radar image observer 20. However, as described above, according to the signal processing device according to the present invention, the constant α can be automatically set without bothering the observer 20. Once the above constant α is set, the following operation is the same as that of the conventional device. That is, the switch 21 connects the contact O and the contact C, and transfers a series of video luminances read from the image memory 12 to the register 13a. The video brightness transferred to the register 13a is converted into brightness through a multiplier 16 and a clipper 15, and then stored in the video memory 1 in the display device 6.
7 and via the scan converter 18.
displayed on the TV monitor 19.

See Figures 2 and 3 for the reason why the radar image displayed on the TV monitor 19 is converted to brightness so as to maximize the probability of detecting an isolated target by the observer 20 through the above operations. I will explain while doing so. FIG. 2 is a one-dimensional representation of the image brightness stored in the image memory 12, and FIG. 3 is a diagram showing the image brightness stored in the image memory 12.
FIG. 3 is a diagram showing the relationship between the coefficient K stored in the image data and the isolated target detection probability. In Figure 2, A is the optimal α value,
B is the average value of the image brightness of the isolated target, and C is the inappropriate α value. Radar images are generally two-dimensional information, but in order to simplify the following explanation, FIG. 2 shows the image brightness distribution in only one of two orthogonal directions. The one-dimensional handling of video information in the following explanation can be easily extended to two-dimensional without impairing the generality of the explanation.

Now, FIG. 2 is a diagram showing a typical example of the brightness distribution of a radar image, where the horizontal axis shows pixel addresses and the vertical axis shows brightness. Here, it is assumed that the isolated target that the observer 20 wants to detect is made up of a series of pixels from address T _S to address T _E , and the other pixels make up the background.
As shown in FIG. 2, the speckle of large image brightness fluctuations seen with the background and isolated targets is a well-known phenomenon, which occurs because the image radar is a coherent system.

If the image brightness obtained by the image radar can be displayed as is, it is obvious that there is a target whose brightness is higher than the background brightness, as shown in FIG. However, as mentioned above, the brightness display range of the display device is not sufficient compared to the brightness range of the radar image, so it is necessary to perform brightness conversion and compress the brightness range of the radar image by the method described above. Now, from the relationship in equation (2), if the value of α is set to the value of Ha in Figure 2, all image brightness greater than the value of Ha will be rounded to the value of Ha, making it difficult to detect isolated targets. It is clear that the probability of

In order to automatically avoid the above-mentioned situation, the video radar signal processing device according to the present invention is configured to be able to set the optimum clip value A from the average value of the target video brightness. That is, α
The value of is determined to be a constant multiple of the average brightness of the isolated target, as shown in the following equation: α=KAp (5) Visual experiment results have shown that the optimal value of α can be determined based on equation (5) above. FIG. 3 shows the relationship between the isolated target detection probability and the value of K based on the results of a visual experiment when the size of n is made to match the size of the isolated target. The horizontal axis is K
, and the vertical axis is the detection probability of an isolated target.

According to the visual experiment, when the size of n matches the size of the isolated target, the relationship shown in Figure 3 holds regardless of the size of the isolated target or the difference in average brightness between the background and the target. can get. The above coefficient K
If K is greater than 2, the probability of detecting an isolated target will not change significantly. Therefore, by setting K to 2 or more, the signal processing device according to the present invention can determine the brightness that maximizes the probability of target detection by the observer. Conversion can be performed automatically.

Here, in many cases, the isolated targets that you want to detect include, for example, a "ship on the sea,""a building in a grassy field,"
The targets are artificial objects that are isolated among natural objects, such as a ``bridge over a river'' (therefore, T _S to _{T E} shown in Figure 2 indicate isolated targets). Therefore, since the observer knows in advance the approximate size of the isolated target (target object), he or she can determine an approximately appropriate size of n in advance, taking into account the resolution (pixel size) of the video radar. can do. However, since the size of an isolated target is not known accurately in advance, the range of addresses (I-n, I+n) for which the average brightness is calculated is the range of addresses where the target exists (T), as shown in Figure 2. _s , T _E ). For example, FIG. 2 shows a case where the former is larger than the latter. Therefore, the calculation result of the average luminance of the target will include an error. However, as shown in Figure 3, there is a wide range of values for K to maximize the probability of detection by an observer, so it is possible to absorb the calculation error of the average brightness by increasing K. Therefore, the value of n, which determines the range of addresses for which the average luminance is to be determined, is not required to be very precise. Note that by setting the coefficient K larger, the background image brightness can be displayed darker on the TV monitor 19. As a result, by setting the coefficient K sufficiently large,
For observer 20, the background image brightness is uniformly black.
It can also be made to look like it is being displayed on the TV monitor 19. However, depending on the observer, there may be cases where it is better to display changes in the brightness of the background as much as possible. In this case, in order to display as much background gradation as possible without deteriorating the detection probability of the target, the constant K should be set to 2 or 3.
It is desirable to set it within the range of .

〔Effect of the invention〕

As explained above, this invention automatically converts the brightness of radar images without the observer's intervention, and displays the radar images on the TV monitor in a way that maximizes the probability that the observer will detect the target. This has the effect of allowing radar images to be obtained in real time.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a diagram showing the brightness distribution of a radar image, FIG. 3 is a diagram showing the relationship between coefficient K and isolated target detection probability, and FIG.
Figure 5 is a diagram for explaining the operation of the video radar.
The figure is a configuration diagram showing a conventional video radar signal processing device. In the figure, 5 is a signal processing device, 6 is a display device, 12 is an image memory, 14 is a divider, 15
is a clipper, 16 is a multiplier, 22 is a maximum value detector, 23 is an average value calculator, and 25 is a multiplier.
Note that the same reference numerals in each figure indicate the same or corresponding parts.

Claims (1)

[Scope of Claims] 1. In a signal processing device for a video radar that converts the brightness of a radar image acquired by a video radar, the signal processing device is configured to convert the brightness of a radar image obtained by a video radar, and the signal processing device is configured to convert the brightness of a radar image obtained by a video radar, wherein the signal processing device is configured to convert the brightness of a radar image obtained by a video radar, and the signal processing device is configured to convert the brightness of a radar image acquired by a video radar, and the signal processing device provides The brightness of the radar image is calculated from the brightness data of the radar image acquired by the video radar, and the image memory stores the brightness of the radar image as numerical data, and the maximum value detects the maximum value of the brightness. an average value calculator that calculates the average brightness of pixels in the vicinity of the detector and the pixel (Piccell) having the maximum brightness over a range set by an observer of the video radar according to the size of the target; a first multiplier that calculates a product A of the average brightness obtained by the average value calculator and a first constant; and a multiplier that calculates the quotient of the product A and a second constant that is the maximum brightness. a second multiplier that calculates a product B of the quotient and the brightness in the image memory, and compares the magnitude relationship between the product B and the second constant, and if the product B is large; and a clipper for converting the value of the product B into the value of the second constant, which is the maximum brightness. 2. The video radar signal processing device according to claim 1, wherein a positive constant in the range of 2 to 3 is used as the first constant.