JP2000003436A - Device and method for recognizing isar picture - Google Patents

Abstract

PROBLEM TO BE SOLVED: To surely identify a desired picture at high speed by extracting it based on a result after collating a detected parameter having a feature with prescribed information which is previously stored in an identifying database. SOLUTION: A silhouette extracting circuit 2 detects the boundary of the picture having a positive inclination, which is obtained by absolute value conversion, and outputs a silhouette picture. An edge extracting part executes a processing for detecting the contour line or the like of an object picture in an ISAR picture. The picture whose edge is extracted is outputted from an output end to the feature quantity extracting circuit 3 of a succeeding stage as the silhouette picture. The feature quantity extracting circuit 3 detects the parameter having the feature such as the length or the like of the object picture, for example, based on the boundary position of the silhouette picture which is outputted from the silhouette extracting circuit. An automatic identifying circuit 4 collates the parameter having the feature, which is detected by the feature quantity extracting circuit 3 with various still pictures stored in the identifying database and outputs its identification result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an ISAR (Inv
erse Synthetic Aperture R
The present invention relates to an ISAR image identification device and an ISAR image identification method for identifying an image.

[0002]

2. Description of the Related Art A conventional ISAR image identification apparatus has, for example, a configuration shown in FIG. In the ISAR image identification device shown in the figure, an ISAR image is input to a still image generation circuit 110, and the output is input to a feature amount measurement circuit 111. The feature quantity measuring circuit 111
Output the image identification result.

Next, the operation of the ISAR image identification device will be described. In the ISAR image identification device shown in FIG. 12, a good image is selected from the ISAR image that changes every moment by visual observation of a human (operator), and the selected image is extracted as a still image by the still imaging circuit 110. The still image extracted by the still image generation circuit 110 is visually checked by a human, and a measurement range such as a length is input to the feature amount measurement circuit 111.

[0004] The feature quantity measuring circuit 111 measures the length or the like from the still image selected by the still imaging circuit 110 within a measurement range designated by a human. Then, based on the measurement result of the feature amount measurement circuit 111, a human compares the identification table with the identification table to identify the ISAR image.

[0005]

However, in the conventional ISAR image identification device having the above-described configuration, a human selects a good image from the constantly changing ISAR image, and the selected still image is selected. You have to enter the measurement range, such as length, which involves human judgment,
There is a problem that the accuracy of the measurement result becomes personal, and it is difficult to always make a constant determination.

[0006] Further, in the conventional ISAR image identification device, it is necessary to select a good image from the ISAR image that changes every moment. Therefore, the operator must always pay close attention to the changing image. However, there is a problem that an excessive burden is imposed on the operator.

[0007] The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide an ISA that changes every moment.
An object of the present invention is to provide an ISAR image identification device and an ISAR image identification method for extracting a good image from R images without human intervention.

[0008]

In order to achieve the above object, the present invention provides an ISAR image identifying apparatus for extracting a desired image from an input ISAR image.
Absolute value converting means for converting an R image into an absolute value to obtain an image having a positive slope, silhouette extracting means for obtaining a silhouette image of the image based on the boundary line of the image having a positive slope, Means for detecting characteristic parameters of the silhouette image based on the boundary position of the silhouette image, means for comparing the characteristic parameters with predetermined information previously stored in an identification database, Means for extracting the desired image based on the
An AR image identification device is provided.

The present invention further comprises means for detecting the maximum value of the feature value of the image having the positive slope, wherein the silhouette extracting means detects a silhouette of the image from the image having the maximum value of the feature value. Get an image.

The present invention further comprises means for comparing the magnitude relationship between the number of pixels of the target image in the ISAR image and a predetermined threshold value, and means for obtaining the characteristic amount of the target image based on the comparison result. And the characteristic amount is set as the characteristic parameter.

Further, the present invention further comprises means for detecting the maximum value of the feature amount of the image having the positive inclination,
Means for comparing the magnitude relationship between the number of pixels of the target image in the AR image and a predetermined threshold value; and means for obtaining a feature amount of the target image based on the comparison result, wherein the silhouette extracting means includes: From the image having the maximum value, a silhouette image of the image is obtained, and the characteristic amount of the target image is set as the characteristic parameter.

Each of the present invention further includes means for extracting a target image having a predetermined feature amount from the ISAR image, means for detecting whether or not the central axis of the target image is rotating, Means for correcting the central axis to a predetermined position when the central axis is rotating, and inputting an image having a constant inclination obtained by the correction to the absolute value converting means.

The present invention also provides an ISAR image identification method for extracting a desired image from an input ISAR image, wherein the absolute value of the ISAR image is obtained to obtain an image having a positive slope. A silhouette extraction step of obtaining a silhouette image of the image based on the boundary line of the image having the positive slope, and a step of detecting characteristic parameters of the image based on the boundary position of the silhouette image. A step of comparing the characteristic parameters with predetermined information previously stored in an identification database, and a step of extracting the desired image based on the result of the comparison.

Further, the above-mentioned ISAR image identification method includes a step of detecting a maximum value of a feature value of the image having the positive slope, and the silhouette extracting step includes a step of extracting the maximum value of the feature value from the image having the maximum value of the feature value. Get a silhouette image of the image.

Further, the above-mentioned ISAR image identification method further comprises a step of comparing the magnitude relation between the number of pixels of the target image in the ISAR image and a predetermined threshold value, and the step of comparing Obtaining a characteristic amount, wherein the characteristic amount is set as the characteristic parameter.

Further, the above-mentioned ISAR image identification method comprises the steps of: detecting a maximum value of a feature amount of the image having a positive slope; and determining a magnitude relationship between the number of pixels of the target image in the ISAR image and a predetermined threshold value. A step of comparing, based on the comparison result, a step of obtaining a feature amount of the target image, wherein the silhouette extraction step obtains a silhouette image of the image from the image having the maximum value, The feature amount of the target image is set as the characteristic parameter.

Each of the above-mentioned ISAR image identification methods includes:
Further, a step of extracting a target image having a predetermined feature amount from the ISAR image, a step of detecting whether or not a central axis of the target image is rotating, and a step of: Correcting the center axis to a predetermined position. In the absolute value conversion step, an image having a constant inclination obtained by the correction is input.

[0018]

Embodiments of the present invention will be described below with reference to the accompanying drawings. Embodiment 1 FIG. First, Embodiment 1 of the present invention will be described. FIG. 1 is a block diagram showing a main configuration of an ISAR image identification device according to the present embodiment (hereinafter, the device will refer to the ISAR image identification device unless otherwise specified). As shown in FIG. 1, the apparatus includes an absolute value detection circuit 1, a silhouette extraction circuit 2, a feature amount extraction circuit 3,
An automatic identification circuit 4 and an identification database 5 are provided.

Next, the operation of the present apparatus will be described. An absolute value detecting circuit 1 shown in FIG. 1 receives a constantly changing ISAR image, and among these images, one ISAR still image is used by the absolute value detecting circuit 1 to determine the absolute value of the "Doppler component". The “inversion image” is inverted by the conversion. This Doppler component is a component that gives the inclination of a target image, for example, an image of a ship or an aircraft, in the ISAR image. Here, only an image having a positive inclination is obtained by absolute value conversion. Also, an inverted image is an image whose characteristic amount has a negative slope.

FIG. 2 is a block diagram showing the internal configuration of the silhouette extraction circuit 2. This silhouette extraction circuit 2
Detects a boundary line of an image having a positive slope obtained by the above-described absolute value conversion, and outputs a silhouette image. That is, an image having a positive slope is input from the absolute value detection circuit 1 to the input terminal 11 in FIG.
The image is binarized. And the image after binarization is
The filter 13 performs smoothing and low-frequency emphasis to suppress noise components.

The edge extracting unit 14 performs a process for detecting a contour line or the like of the target image in the ISAR image. In general, at the boundary portion of an image, since the density value changes rapidly, a pixel serving as an edge candidate is compared with its neighboring pixel value, and the pixel value of the neighboring pixel is equal to or larger than a predetermined value. If it indicates a change, the pixel serving as the edge candidate is defined as an edge. The image on which the edge extraction has been performed in this manner is output as the silhouette image as a silhouette image.
5 is output to the feature extraction circuit 3 at the next stage.

The characteristic amount extraction circuit 3 detects characteristic parameters such as the length of the target image based on the boundary position of the silhouette image output from the silhouette extraction circuit 2. The automatic identification circuit 4 collates the characteristic parameters detected by the characteristic amount extraction circuit 3 with various still images stored in the identification database 5, and identifies the identification result (I
SAR image identification result).

The above-mentioned identification database 5 includes:
More specifically, a three-view drawing (still image) of various ships and their feature amounts (for example, lengths and the like) are accumulated, and the automatic identification circuit 4 detects the target image detected by the feature amount extraction circuit 3. Are compared with the information in the identification database 5.

The ISAR image identification device according to the present embodiment performs the above-described processing on all the input ISAR images, thereby extracting the feature amount of the target image and determining the characteristic amount of the operator. Automatic identification of target images independent of.

As described above, according to the present embodiment, characteristic parameters of an extracted silhouette image are detected based on the boundary position of the image, and the characteristic parameters are stored in advance in an identification database. By collating with a still image or the like, desired image identification can be performed quickly and reliably for all the input ISAR images.

Embodiment 2 FIG. Hereinafter, Embodiment 2 of the present invention will be described. FIG. 3 shows the IS according to the present embodiment.
FIG. 2 is a block diagram illustrating a main configuration of an AR image identification device (hereinafter, referred to as a device). In the device shown in FIG.
The same components (absolute value detection circuit, silhouette extraction circuit, feature quantity extraction circuit, automatic identification circuit, identification database) as those of the apparatus according to the first embodiment shown in FIG. , And their description is omitted.

In the first embodiment, the I that changes every moment
For one still image in the SAR image, the inverted image is inverted by the absolute value detection circuit 1 and then converted into a silhouette image by the silhouette extraction circuit 2. In the present embodiment, as shown in FIG. The maximum value hold circuit 6 is inserted before the silhouette extraction circuit 2.

The maximum value hold circuit 6 shown in FIG. 3 includes, for example, the ISAR image in the absolute value detection circuit 1 in the ISAR image.
The maximum value of the characteristic amount unique to the ship, such as the length and height of the ship, is detected. Here, the maximum value does not indicate the amount of continuation of the same pixel in a certain image, but means an image in which the largest feature amount is displayed in an ever-changing ISAR image. In other words, the object to be held (held) by the maximum value hold circuit 6 is not a specific portion in one image but an entire ISAR image that changes every moment.

FIG. 4 is a diagram schematically showing a procedure of image processing according to the present embodiment. (A) of FIG.
An image is an image that changes every moment according to the flow of time indicated by t in the figure. (B) is an image when the absolute value detection circuit 1 converts the absolute value. As described above, here, only the image having a positive inclination is obtained by the absolute value conversion.

With respect to the image (b) that has been converted into an absolute value in this manner, the maximum value hold circuit 6 applies a target image (here,
The maximum value of the characteristic amount unique to the ship) is detected. Then, the silhouette extraction circuit 2 detects a boundary line of an image having a positive slope in the image having the maximum value of the feature amount in the same manner as in the first embodiment, and based on this, detects the silhouette image. (D) is output.

As described above, in the present embodiment, the maximum value of the target image in the constantly changing ISAR image is detected, and an image to be processed after the silhouette extraction circuit is selected from the plurality of ISAR images. The IS
Since the number of AR still images can be reduced, the overall processing time can be significantly reduced.

Embodiment 3 FIG. Hereinafter, Embodiment 3 of the present invention will be described. FIG. 5 shows the IS according to the present embodiment.
FIG. 2 is a block diagram illustrating a main configuration of an AR image identification device (hereinafter, referred to as a device). In the device shown in FIG.
The same components as those of the apparatus according to the first embodiment shown in FIG. 1 are denoted by the same reference numerals, and description thereof will be omitted.

The apparatus shown in FIG. 5 has a configuration in which a threshold determination circuit 7 and a threshold database 8 are inserted after the silhouette extraction circuit 2 in the apparatus according to the first embodiment shown in FIG. Among these, the threshold determination circuit 7
In the target image, for example, in an image of a ship or the like,
The number of pixels of an image having the same pixel value continuously and its distribution are calculated. In the threshold database 8, thresholds are set for the threshold determination circuit 7 to determine whether the target image is a specific image.

By comparing the number of pixels of the target image obtained by the above calculation and the distribution thereof with data in the threshold database 8, it is possible to determine which target image corresponds to which part of the target image. Therefore, when taking a ship etc. as an example,
Ship-specific features such as the length of the bridge and its position (for example, position with respect to the bow) are obtained. The result obtained here is output to the feature extraction circuit 3 at the next stage.

As described above, according to the present embodiment, the number of pixels in an ever-changing ISAR image,
By detecting the magnitude relationship with the threshold value in the threshold value database and obtaining the feature amount of the target image based on the result, an image to be processed after the silhouette extraction circuit can be selected from a plurality of ISAR images. Since the number of ISAR still images to be processed can be reduced, the overall processing time can be reduced.

Embodiment 4 FIG. FIG. 6 is a block diagram showing a main configuration of an ISAR image identification apparatus (hereinafter, referred to as an apparatus) according to the present embodiment. In the device shown in the figure, the same components as those of the device according to the third embodiment shown in FIG. 5 are denoted by the same reference numerals, and the description thereof is omitted here.

In the apparatus according to the present embodiment shown in FIG.
As in the device according to the second embodiment shown in FIG. 3, a maximum value hold circuit 6 is arranged at a stage subsequent to the absolute value detection circuit 1, and the processing result is input to the silhouette extraction circuit 2. Embodiment 3 shown in FIG. 5 as a whole
And a device configuration in which the maximum value hold circuit 6 is added to the device according to (1). Therefore, the threshold value judgment circuit 7 and the threshold value database 8 are inserted after the silhouette extraction circuit 2.

Also in the present embodiment, the maximum value holding circuit 6 uses the ISAR
The maximum value of the feature amount of the target image in the image is detected. Also,
The threshold value judgment circuit 7 and the threshold value database 8
The magnitude relationship between the number of pixels of the AR image and the threshold in the threshold database is detected, and the feature amount of the target image is obtained from the result.

With this configuration, the apparatus according to the present embodiment detects the maximum value of the target image in the constantly changing ISAR image and the magnitude relationship between the number of pixels and the threshold value in the ISAR image. Then, the feature amount of the target image is obtained in detail, and as a result, an image to be processed after the silhouette extraction circuit can be selected from a plurality of ISAR images.
The number of ISAR still images to be processed can be reduced, and the overall processing time can be further reduced.

Embodiment 5 FIG. 7 is a block diagram showing a main configuration of an ISAR image identification apparatus (hereinafter, referred to as an apparatus) according to the present embodiment. In the apparatus shown in the figure, the same components as those of the apparatus according to the first embodiment shown in FIG. 1 are denoted by the same reference numerals, and the description thereof is omitted here.

The device shown in FIG. 7 is different from the device according to the first embodiment shown in FIG. 1 in that a pitch motion correcting circuit 9 is arranged at a stage preceding the absolute value detecting circuit 1.
Therefore, here, the ISAR image is directly input to the pitch motion correction circuit 9, and the circuit performs a correction process described later. The corrected image is shown in FIG.
The signal is input to a device having the same configuration as the device according to the first embodiment, that is, a device starting from the absolute value detection circuit 1.

FIG. 8 is a diagram schematically showing a procedure of image processing (more specifically, pitch motion correction processing) according to the present embodiment. (A) of the figure shows an ISAR image, which is an image that changes every moment according to the flow of time indicated by t in the figure. The pitch motion correction circuit 9 shown in FIG.
When the central axes P ₁ and P _{2 of the} target image having the characteristic amount of a ship or the like are rotating, as shown in ( ₂ ), a process of rotating those axes to correct the axes to a normal position is performed. Here, the central axes P ₁ and P ₂ are rotated by angles θ ₁ and θ ₂ , respectively.

As a result of the above pitch motion correction processing, an image having a constant inclination shown in FIG. 8B is obtained, and the image is converted to an absolute value by the absolute value detection circuit 1 and the image is converted to an absolute value as shown in FIG. As shown in (), only an image having a positive inclination is obtained. Subsequent processing is the same as that of the device according to the first embodiment.

As described above, according to the present embodiment, the ISA having a constant inclination is performed by the pitch motion correction processing.
Since an R still image is obtained, the extraction accuracy in the feature amount extraction circuit can be significantly improved.

Also, by this pitch motion correction processing,
Since the inclination of the ISAR still image is all constant, the amount of data relating to characteristic parameters and external features of the target image is reduced, and the size of the identification database can be reduced.

Embodiment 6 FIG. FIG. 9 is a block diagram showing a main configuration of an ISAR image identification device (hereinafter, device) according to the present embodiment. In the apparatus shown in the figure, the same components as those of the apparatus according to the second embodiment shown in FIG. 3 are denoted by the same reference numerals, and the description thereof is omitted here.

The device shown in FIG. 9 is provided before the absolute value amplifier circuit 1 of the device according to the second embodiment shown in FIG.
The pitch motion correction circuit 9 according to the fifth embodiment shown in FIG. That is, also in the present embodiment, similarly to Embodiment 5 described above, an image having a constant inclination, which has been subjected to pitch motion correction processing by the pitch motion correction circuit 9,
The absolute value is converted into an absolute value in the absolute value detection circuit 1, and the image is
It is input to the maximum value hold circuit 6.

The maximum value hold circuit 6 detects the maximum value of the feature amount (for example, the length and height of a ship) of the target image in the ISAR image that has been converted into the absolute value by the absolute value detection circuit 1. Thereafter, the same processing as in the second embodiment is performed.

As described above, according to the present embodiment, by inserting the pitch motion correction circuit in the preceding stage of the absolute value amplification circuit, the inclination of the characteristic amount, that is, the image of the ship or the like in the ISAR image has Since an ISAR still image with a constant inclination is obtained, the measurement accuracy of the maximum value in the constantly changing ISAR image can be improved, and the extraction accuracy of the feature amount extraction circuit can be improved.

Embodiment 7 FIG. Hereinafter, a seventh embodiment of the present invention will be described. FIG. 10 is a diagram showing an I-mode according to the present embodiment.
FIG. 2 is a block diagram illustrating a main configuration of an SAR image identification device (hereinafter, referred to as a device). In the device shown in the figure, the same components as those of the device according to the third embodiment shown in FIG. 5 are denoted by the same reference numerals, and the description thereof is omitted here.

As shown in FIG. 10, the device according to the present embodiment is provided at a stage preceding the absolute value detecting circuit 1 of the device according to the third embodiment shown in FIG. 5 is inserted. That is, an image having a constant inclination, which has been subjected to pitch motion correction processing in the pitch motion correction circuit 9, is converted into an absolute value in the absolute value detection circuit 1, and is converted into a threshold value by the threshold extraction circuit 7 via the silhouette extraction circuit 2. Is input to

The threshold value judging circuit 7 calculates the number of pixels of an image having the same pixel value continuously in the target image and its distribution, as in the third embodiment. It should be noted that the threshold value is set in the threshold value database 8 when the threshold value determining circuit 7 determines whether or not the target image is a specific image. Thereafter, the same processing as in the third embodiment is performed.

With the apparatus according to the present embodiment having such a configuration, it is possible to obtain an ISAR still image having a constant feature amount inclination, and furthermore, to determine the magnitude between the number of pixels in the ever-changing ISAR image and the threshold value. Since the relationship can be detected accurately, the measurement accuracy of the number of pixels in the ever-changing ISAR image and the extraction accuracy of the feature amount extraction circuit can be improved.

Embodiment 8 FIG. Hereinafter, an eighth embodiment of the present invention will be described. FIG. 11 is a diagram showing an I-mode according to the present embodiment.
FIG. 2 is a block diagram illustrating a main configuration of an SAR image identification device (hereinafter, referred to as a device). In the device shown in the figure, the same components as those in the device according to the fourth embodiment shown in FIG. 6 are denoted by the same reference numerals, and the description thereof is omitted here.

As shown in FIG. 11, the device according to the present embodiment is arranged at a stage preceding the absolute value detection circuit 1 of the device according to the fourth embodiment shown in FIG. 5 is inserted. Here, similarly to the fifth embodiment, an image having a constant inclination, which has been subjected to pitch motion correction processing by the pitch motion correction circuit 9,
The absolute value is converted into an absolute value in the absolute value detection circuit 1, and the image converted into the absolute value is input to a maximum value hold circuit 6 provided in a subsequent stage.

Thereafter, similarly to the apparatus according to the fourth embodiment, the maximum value of the feature amount is detected by the maximum value hold circuit 6 and is input to the threshold value judgment circuit 7 via the silhouette extraction circuit 2. As a result, a magnitude relationship between the number of pixels of the constantly changing ISAR image and a threshold in the threshold database is detected, and based on the result, a feature amount of the target image is obtained.

As described above, according to the present embodiment, the pitch motion correction circuit is inserted before the absolute value amplification circuit to obtain an ISAR still image having a constant feature amount inclination, and thereafter, every moment. By detecting the maximum value of the target image in the changing ISAR image and the magnitude relationship between the number of pixels and the threshold for the ISAR image, and obtaining the feature amount of the target image in detail,
The accuracy of measurement of the number of pixels in the ISAR image that changes every moment can be improved, and the accuracy of extraction by the feature amount extraction circuit can be improved.

[0058]

As described above, according to the present invention,
By extracting a desired image based on the result of collating the detected characteristic parameters with predetermined information previously stored in the identification database, all the input IS
For an AR image, desired image identification can be performed quickly and reliably.

According to another aspect of the present invention, the maximum value of the target image in the ISAR image is detected, and an image to be processed after the silhouette extraction circuit is selected from among the plurality of ISAR images, whereby the processing target The number of ISAR still images can be reduced, and the overall processing time can be greatly reduced.

According to another aspect of the present invention, means for comparing the magnitude relationship between the number of pixels of the target image in the ISAR image and a predetermined threshold value, and means for obtaining the characteristic amount of the target image based on the comparison result Since the number of ISAR still images for which the characteristic amount is set as the characteristic parameter can be reduced, the overall processing time can be reduced.

According to another aspect of the present invention, the maximum value of the feature value of an image having a positive slope is detected, and the result of comparing the magnitude relationship between the number of pixels of the target image in the ISAR image and a predetermined threshold value is obtained. Originally, the feature amount of the target image is obtained, and a silhouette image of the image is obtained from the image having the maximum value.
By using the characteristic amount of the target image as the characteristic parameter, an image to be processed after the silhouette extraction circuit can be selected from a plurality of ISAR images, and the number of ISAR still images to be processed can be reduced. The overall processing time can be further reduced.

According to another invention, when the center axis of the target image is rotating, the center axis is corrected to a predetermined position, and the image having a constant inclination obtained by this correction is converted into an absolute value. ISAR with a constant inclination
Since a still image can be obtained, the extraction accuracy in the feature amount extraction circuit can be significantly improved, and further, the amount of data relating to the characteristic parameters and external features of the target image decreases, and the size of the identification database becomes smaller Small enough.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a main configuration of an ISAR image identification device according to Embodiment 1 of the present invention.

FIG. 2 is a block diagram illustrating an internal configuration of a silhouette extraction circuit.

FIG. 3 is a block diagram showing a main configuration of an ISAR image identification device according to a second embodiment;

FIG. 4 is a diagram schematically showing a procedure of image processing according to a second embodiment.

FIG. 5 is a block diagram illustrating a main configuration of an ISAR image identification device according to a third embodiment;

FIG. 6 is a block diagram illustrating a main configuration of an ISAR image identification device according to a fourth embodiment;

FIG. 7 is a block diagram illustrating a main configuration of an ISAR image identification device according to a fifth embodiment;

FIG. 8 is a diagram schematically showing a procedure of image processing (pitch motion correction processing) according to a fifth embodiment.

FIG. 9 is a block diagram illustrating a main configuration of an ISAR image identification device according to a sixth embodiment;

FIG. 10 is a block diagram illustrating a main configuration of an ISAR image identification device according to a seventh embodiment;

FIG. 11 is a block diagram showing a main configuration of an ISAR image identification device according to an eighth embodiment.

FIG. 12 is a block diagram showing a configuration of a conventional ISAR image identification device.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Absolute value detection circuit, 2 ... Silhouette extraction circuit, 3 ... Feature amount extraction circuit, 4 ... Automatic identification circuit, 5 ... Identification database, 6 ... Maximum value hold circuit, 7 ... Threshold judgment circuit, 8
... Threshold database, 9 ... Pitch motion correction circuit, 11 ...
Input end, 12 binarization unit, 13 filter, 14 edge extraction unit, 15 output end

Claims (12)

[Claims] 1. An ISAR image identification device for extracting a desired image from an input ISAR image, comprising: an absolute value converting means for converting the ISAR image into an absolute value to obtain an image having a positive slope; A silhouette extracting unit that obtains a silhouette image of the image based on a boundary line of the image having: a unit that detects a characteristic parameter of the image based on a boundary position of the silhouette image; An ISAR image identification apparatus, comprising: means for comparing various parameters with predetermined information previously stored in an identification database; and means for extracting the desired image based on the result of the comparison. 2. The ISA according to claim 1, wherein the identification database stores, as the predetermined information, data obtained by digitizing physical feature amounts of the target image.
R image identification device. 3. The ISAR image identification device according to claim 1, wherein said absolute value conversion means converts the Doppler component of the ISAR image into an absolute value. 4. A method for detecting a maximum value of a feature amount of an image having a positive slope, wherein the silhouette extracting unit converts a silhouette image of the image from the image having the maximum value of the feature amount. The ISAR image identification device according to any one of claims 1 to 3, wherein the ISAR image identification device is obtained. 5. A means for comparing a magnitude relationship between the number of pixels of the target image in the ISAR image and a predetermined threshold value, and a means for obtaining a characteristic amount of the target image based on the comparison result. The ISAR image identification device according to any one of claims 1 to 3, further comprising: using the characteristic amount as the characteristic parameter. 6. A means for detecting a maximum value of a feature amount of the image having a positive slope, a means for comparing a magnitude relationship between the number of pixels of a target image in the ISAR image and a predetermined threshold value, Means for obtaining a feature amount of the target image based on the comparison result, wherein the silhouette extracting means obtains a silhouette image of the image from the image having the maximum value, and further includes a feature of the target image. 4. The ISAR image identification device according to claim 1, wherein an amount is used as the characteristic parameter. 7. A means for extracting a target image having a predetermined feature amount from the ISAR image; a means for detecting whether or not a central axis of the target image is rotating; Means for correcting the center axis to a predetermined position, and inputting an image having a constant inclination obtained by the correction to the absolute value conversion means. 7. The ISAR image identification device according to any one of 6. 8. An ISAR image identification method for extracting a desired image from an input ISAR image, an absolute value converting step of converting the ISAR image to an absolute value to obtain an image having a positive slope, A silhouette extraction step of obtaining a silhouette image of the image based on a boundary line of the image having; and a step of detecting characteristic parameters of the image based on a boundary position of the silhouette image; An ISAR image identification method, comprising the steps of: comparing various parameters with predetermined information previously stored in an identification database; and extracting the desired image based on the result of the comparison. 9. The method according to claim 1, further comprising the step of detecting a maximum value of a feature amount of the image having the positive slope, wherein the silhouette extracting step converts a silhouette image of the image from the image having the maximum value of the feature amount. 9. The ISAR image identification method according to claim 8, wherein the ISAR image is obtained. 10. A step of comparing a magnitude relationship between the number of pixels of the target image in the ISAR image and a predetermined threshold value; and a step of obtaining a feature amount of the target image based on the comparison result. The ISAR image identification method according to claim 8, wherein the feature amount is set as the characteristic parameter. 11. A step of detecting a maximum value of a feature amount of the image having a positive slope; a step of comparing a magnitude relationship between the number of pixels of the target image in the ISAR image and a predetermined threshold value; Obtaining a feature amount of the target image based on the comparison result, wherein the silhouette extracting step obtains a silhouette image of the image from the image having the maximum value, and further includes a feature of the target image. 9. The ISAR image identification method according to claim 8, wherein an amount is used as the characteristic parameter. 12. A step of extracting a target image having a predetermined characteristic amount from the ISAR image; a step of detecting whether or not a central axis of the target image is rotating; Correcting the center axis to a predetermined position, and inputting an image having a constant inclination obtained by the correction in the absolute value converting step. The method for identifying an ISAR image according to any one of the eleventh to eleventh aspects.