JP2011112422A - Image processor and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor and an image processing method for a radar image and a true track image obtained by the true track processing of the radar image which removes a static object while controlling an effect of the fluctuation of an echo without obtaining external information about the static object. <P>SOLUTION: The image processor for generating an image by radar signal processing includes a calculation means for obtaining the image as a product set of the radar image and the true track image obtained by true track processing of the radar image. Also, the radar image and the true track image are binarized and the logical product of the both images can be obtained as the product set. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

The present invention relates to a radar image and an image processing apparatus and an image processing method for performing arithmetic processing on a true track image obtained by performing true track processing on the radar image and outputting an image.

In a capture and tracking device that captures and tracks moving targets, a stationary target such as land is mistakenly determined as a target for capture and tracking where a moving target such as another ship should be targeted. There was a problem. This is because the radar reception signal is mixed with the reception signal reflected by a stationary target such as land.

When the land is captured and tracked, a phenomenon called “transfer” appears on the screen as if other ships are sailing on the land. Such a phenomenon is a major obstacle in navigation.

In order to solve the above problem, the radar reception signal received by the radar is present on the radar image obtained by imaging from the polar coordinate format to the orthogonal coordinate format, or on the true track image obtained by performing the true track processing on the radar image. It is effective to remove the image of the stationary target.

For example, in Patent Document 1, in a comparison of radar images obtained in the current scan and the previous scan with respect to a radar image, a moving target exists at coordinates where there is a change in the image, and a stationary object exists at coordinates where there is no change. This shows a technique in which a moving target is extracted by removing a stationary target image from a radar image by determining that a target exists. Scan here refers to the period during which the antenna moves once. The comparison is a simple logical operation.

Patent Document 2 shows a technique for removing an image of a stationary target by specifying the position of a stationary target such as a land on the radar image based on chart information in the same range as the radar image. Yes.

JP-A-8-15408 Japanese Patent Laid-Open No. 11-23707

However, a simple image comparison as in the method of Patent Document 1 cannot completely eliminate a stationary target because the influence of echo fluctuation cannot be eliminated. Echo fluctuation means a state in which, when a stationary target is continuously observed on a radar image, the position of a stationary target that should not move slightly changes, for example. Causes include external factors such as the influence of distance resolution on signal processing and fluctuations in the radar apparatus itself. When affected by the fluctuation of the echo, there is a problem that the image of the contour portion of the stationary target is erroneously left as a moving target.

In addition, the method of Patent Document 2 has a problem that it cannot be incorporated in a device having a small storage area because it occupies a large area on the storage area by having external information such as chart information in the device.

An object of the present invention is to provide an image processing apparatus and an image processing method for removing a stationary target while suppressing the influence of echo fluctuations without acquiring external information about the stationary target.

An image processing apparatus according to the present invention is an image processing apparatus that generates an image under radar signal processing, and is a product set of a radar image and a true wake image obtained by performing true wake processing on the radar image. An arithmetic means for obtaining an image is provided.

Further, the radar image and the true track image are binarized, and a logical product of both is obtained as the product set.

The image processing method according to the present invention is characterized in that the image is obtained as a product set of a radar image and a true track image obtained by a true track process performed on the radar image.

According to the present invention, a radar image and a true track image obtained by performing true track processing on the radar image are subjected to relatively easy image processing using a small memory area without acquiring external information regarding a stationary target. Thus, it is possible to output an image from which a stationary target is removed while suppressing the influence of echo fluctuation.

In addition, the radar image and the true track image obtained by performing the true track processing on the radar image are binarized and then subjected to image processing, thereby further reducing the memory area to be used and reducing the stationary target. Can be output.

It is a block diagram of an image processing device concerning this embodiment. It is a flowchart which shows the flow of the logical operation in this embodiment. It is a flowchart which shows the flow of the logical operation in this embodiment. It is the figure which modeled the flow of the image processing by this embodiment, and its effect. It is a figure which shows a response | compatibility with the pixel number j and a coordinate position about the binarized image whose number of pixels of the horizontal direction on a screen is X and whose number of vertical pixels is Y.

Next, an embodiment of an image processing apparatus according to the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the image processing apparatus according to this embodiment. FIG. 2 is a flowchart showing the calculation flow in the first image calculation unit 103 of the image processing apparatus according to this embodiment, and FIG. 3 is a flowchart showing the calculation flow in the second image calculation unit 107. FIG. 4 is a diagram for explaining the processing steps and effects of the present embodiment by taking a radar image and a true track image of a radar mounted on a ship as an example. FIG. 5 is a diagram for explaining a method for designating pixel positions on an image in the image processing according to the present embodiment.

In FIG. 1, in the image processing apparatus according to the present embodiment, a radar image and a true wake image obtained by performing a true wake process on the radar image are input to a first image calculation unit 103 arranged in the first stage. Here, the true track processing means image processing for storing and displaying the movement amount of the other target in an absolute position regardless of the position of the radar serving as the observation point. On the other hand, the wake process in which the movement amount of the other target is stored and displayed in the relative position with the position of the radar serving as the observation point as a reference is referred to as a relative wake process.

The output of the first image calculation unit 103 is input to the image storage unit 105 and the second image calculation unit 107 arranged in the final stage. The output of the image storage unit 105 is input to the second image calculation unit 107 together with the output of the first image calculation unit 103. The output of the second image calculation unit 107 is input to the acquisition and tracking device together with the radar image.

The first image calculation unit 103 performs a logical operation on the radar image and the true track image. The image storage unit 105 stores the image output from the first image calculation unit 103. The second image calculation unit 107 performs a logical calculation of the image output by the first image calculation unit 103 and the image stored by the image storage unit 105 during the past scan.

The radar image and the true track image are binarized images. The radar image is binarized when the radar apparatus images the radar reception signal into a radar image. The binary here is the presence or absence of the target, and based on an appropriate threshold determined for the magnitude of the received power of the radar received signal, the target exists if it is above the threshold, and the target is below the threshold. It is distinguished as not existing.

Assume that the binarized radar image is E _n (j) and the true track image is T _n (j). The coordinates where the target exists are E _n (j) = 1, and the coordinates where the target does not exist are E _n (j) = 0. Similarly, in the true track image obtained by performing the true track processing on the radar image, the coordinates where the true track exists are T _n (j) = 1, and the coordinates where the true track does not exist are T _n (j) = 0.

Here, n is an integer representing the number of scans of the radar transmission / reception signal. The variable j is a pixel number that designates the coordinate position of the pixel forming the image in the orthogonal coordinate system. FIG. 5 shows the correspondence between the pixel number j and the coordinate position for the binarized image. The image shown in FIG. 5 has X in the horizontal direction on the screen and Y in the vertical direction. In FIG. 5, the pixel 501 shown black is E _n (j) = 1 or T _n (j) = 1, and the pixel 503 shown white is E _n (j) = 0 or T _n (j) = 0. It is. When the number of pixels of the image is M, M = X × Y and j is 0 <j <M.

The radar image E _n (j) and the true track image T _n (j) are input to the first image calculation unit 103. FIG. 4A shows a radar image E _n (j) of a radar mounted on a ship, and FIG. 4B shows a true track image T _n (j). On the radar image of FIG. 4A, there are images of other ships and land images. In addition, on the true track image of FIG. 4B, the movement amount of the other ship in a continuous time is displayed.

The first image calculation unit 103 extracts a true track of the moving target by removing the echo of stationary target from the true track image T _n. The true track of the moving target exists at the coordinates where there is a true track (T _n (j) = 1) and there is no echo (E _n (j) = 0). Thus, the first image calculation unit 103, the logical operation of the formula (1) for the pixels in the same coordinate positions in the radar image E _n and the true track image T _n. The result of the logical operation is defined as a first extracted image _Pn . The flow of logic operation in the first image calculation unit 103 is shown as a first image calculation in the flowchart of FIG. FIG. 2 shows a flow of processing in which the logical operation of S201 is repeated the same number of times as the number of pixels of the image.

The coordinates of P _n (j) = 1 are coordinates determined by the first image calculation unit 103 as a result of Expression (1) that only the true track of the moving target exists. However, an image of a stationary target such as land remains at the coordinates of P _n (j) = 1 due to the influence of echo fluctuation. FIG. 4C is a first extracted image that is output after the first image calculation is performed on the radar image of FIG. 4A and the true track image of FIG. 4B by the first image calculation unit. In FIG. 4 (C), the contour portion of the land in particular is erroneously extracted together with the true track of the other ship.

Therefore, the second image calculation unit 107 suppresses the influence of echo fluctuation. The first extracted image P _n (j) output from the first image calculation unit 103 is input to the second image calculation unit 107 and the image storage unit 105. The second image calculation unit 107 removes the image of the contour portion of the stationary target from the first extracted image P _n (j), and extracts only the true track newly added in the current scan. The image storage unit 105 stores the first extracted image P _n (j) that is the target of the logical operation in the second image calculation unit 107 over a plurality of times during the past scan.

In addition to the first extracted image P _n (j) of the current scan, the first extracted image P _n−1 (j) stored in the image storage unit 105 in the previous scan is input to the second image calculation unit 107. The second image calculation unit 107 uses the first extracted image P _n (j) in the current scan and the pixel at the same coordinate position in the _first extracted image P _n−1 (j) stored in the image storage unit 105 in the previous scan. Perform logical operations. FIG. 4 (X) shows a first extracted image stored in the previous scan by the image storage unit 105 during the previous scan. On the image shown in FIG. 4 (X), there are past wakes and contour portions of the remaining land.

An image of a stationary target that has been extracted as a true track of a moving target due to the influence of echo fluctuations moves slightly every scan on the screen. However, unlike the case where the target itself moves, the movement often moves a plurality of adjacent pixels irregularly while staying within a certain limited narrow range. Therefore, in the first extracted image P _n (j) and the first extracted image P _n−1 (j) of the previous scan, a plurality of adjacent coordinates are not included in the image of the stationary target extracted by the fluctuation of the echo. When moving to a rule, the same coordinates often appear.

In addition, the true track newly added in the current scan exists in the first extracted image P _n (j) of the current scan (P _n (j) = 1), while the first extracted image P _n− of the previous scan. ₁ is not present (P _n-1 (j) = 0). Therefore, the second image calculation unit 107 performs a logical calculation of Expression (2) on the first extracted image P _n (j) and the first extracted image P _n−1 (j) of the previous scan. The result of the logical operation is set as the second extracted image Q _n (j).

Furthermore, in consideration of the echo fluctuation characteristic that the position of the stationary target moves slightly on the screen for each scan, the second image calculation unit 107 is the image storage unit 105 as a target of the logical calculation in Expression (2). Is added with the first extracted image P _nk (j) stored for k times at the time of the past scan, and the logical operation of Expression (3) is performed. By the logical operation of Expression (3), the image of the stationary target appearing at the same coordinates for a plurality of times in the first extracted image is removed. The flow of the logical operation in the second image calculation unit 107 is shown in the flowchart of FIG. 3 as the second image calculation. In FIG. 3, the flow of a process of repeating the logical operation of S301 for the first extracted image P _n−k (j) stored for k times at the time of the past scan as many times as the number of pixels of the image. Show.

The coordinates of Q _n (j) = 1 are coordinates determined by the second image calculation unit 107 that there is no still target image and only the newly added true wake is present in the current scan. The second image calculation unit 107 removes the contour portion of the land that has remained in the first extracted image of FIG. 4C, and extracts the second wake of the moving target added in the current scan. The extracted image is shown in FIG.

The second extracted image is output from the image processing apparatus and input to the capture and tracking apparatus 109 connected to the image processing apparatus. When the acquisition and tracking device 109 performs acquisition and tracking processing on the input radar image, the acquisition and tracking device 109 uses the second extracted image to determine an acquisition and tracking target. As a method of using the second extracted image, for example, it is used that when the acquisition tracking target is determined, the acquisition tracking target is narrowed down around the true track newly added in the current scan. For example, after performing normal capture tracking processing on the true track image input to the capture tracking device 109 in advance and finally determining the capture tracking target, Q _n (j) on the second extracted image. There is a method in which a candidate closest to the coordinate at which = 1 is determined as an acquisition and tracking target.

The acquisition and tracking device 109 can perform acquisition and tracking processing only within a certain range around the coordinates where Q _n (j) = 1 in the second extracted image. For example, as shown in FIG. 4 (Z), searching for a capture tracking target limited to a range close to the coordinates where Q _n (j) = 1 on the second extracted image, rather than searching on the entire screen. However, the amount of calculation is reduced and the time required for the acquisition and tracking process is shortened.

As described above, the image processing apparatus according to the present embodiment removes the still target image from the radar image and the true track image. When removing an image of a stationary target, it is possible to output an image in which only a moving target is correctly extracted by suppressing the influence of echo fluctuations and removing the contour of the stationary target.

The output image is input to the capture and tracking device, and the capture and tracking device can determine the capture and tracking target after the stationary target is excluded from the capture and tracking target candidates. As a result, erroneous tracking to a stationary target can be prevented. Furthermore, “transfer” to the land, which has been a problem in the past, can be prevented.

Since the image processing apparatus according to the present embodiment performs a logical operation on the binarized image and outputs an extracted image, the amount of calculation and the amount of data to be stored in the image extraction process are small. Therefore, the memory used can be reduced.

In addition, the image processing apparatus according to the present embodiment performs a relatively simple logical operation on a radar image and a true track image provided in a general radar apparatus and outputs an extracted image. The addition of is not necessary. Therefore, it can be easily provided as an apparatus for removing a stationary target image and extracting a moving target.

In the image processing device according to the above-described embodiment, the output extracted image is input to the capture tracking device. However, the image extraction device is not limited to the capture tracking device, and can be input to other devices to perform further image processing. Needless to say, the method of using the output extracted image is not limited to this embodiment.

The image storage unit 105 in the present embodiment is not limited to a storage device, and can be handled by an element such as a delay device. When storing the first extracted image P _n (j) over k times at the time of past scanning, the number k can be freely determined after considering the degree of echo fluctuation, memory capacity, and the like.

Further, as shown in the present embodiment, the present invention is not limited to a configuration in which a radar image or a true track image is converted into a binarized image, and the coordinate system, the configuration of the array, and the like are not limited.

Therefore, the range of the image to be subjected to image processing does not necessarily cover the entire radar image or true track image, and it is also possible to target only a part of the image.

Furthermore, in a fixed-point observation radar that is fixed to a lighthouse or the like and monitors the sea,
Since the radar itself is stationary, the wake image in which the wake processing is performed on the radar image exhibits the same body regardless of whether the wake processing is the true wake processing or the relative wake processing. When relative track processing is applied to a fixed-point observation radar image, the movement amount of the other target is stored at a relative position with respect to the stationary fixed-point observation radar, that is, the absolute position, and thus the same as the true track processing image. This is because an image can be obtained. Therefore, in the fixed point observation radar, even if the wake processing is not true wake processing, the same effect is obtained by performing the same image processing as in the present embodiment on the radar image and the wake image obtained by performing the wake processing on the radar image. It goes without saying that can be obtained.

As described above, the present invention is not limited to the above-described embodiments, and various configurations of the embodiments are possible within the scope of the present invention, and any improvement may be applied to all or some of the components.

DESCRIPTION OF SYMBOLS 101 Image processing apparatus 103 1st image calculating part 105 Image memory | storage part 107 2nd image calculating part 109 Acquisition tracking apparatus 501 The coordinate in which a target exists in a radar image, The coordinate 503 in which a wake exists in a true track image Coordinates where the target does not exist, coordinates where the wake does not exist in the true wake image

Claims (3)

An image processing device for generating an image under radar signal processing,
An image processing apparatus comprising: a calculation unit that obtains the image as a product set of a radar image and a true track image obtained by performing true track processing on the radar image. The image processing apparatus according to claim 1.
The image processing apparatus, wherein the radar image and the true track image are binarized and a logical product of the two is obtained as the product set. An image processing method, wherein the image is obtained as a product set of a radar image and a true track image obtained by a true track process performed on the radar image.