JPH0792259A - Target classification apparatus - Google Patents

Abstract

PURPOSE:To precisely classify the generation source of acquired information by a method wherein the feature of a moving object is extracted from information by a Doppler radar and the feature is collated with a feature at every object already classified, and stored. CONSTITUTION:Information which is acquired by a radar part 20 is sent to an information analysis part 21, the information is analyzed by, e.g. a high-speed Fourier transform (FFI processing), an analysis result is sent to a feature extraction part 22, and the feature of the acquired information by a Doppler radar is changed to a form which can be collated with a feature at every object classified and stored in advance. A feature collation part 23 collates the feature of the acquired information by the Doppler radar with the feature at every object classified and stored in advance, the ratio of their coincidence is found, and what the generation source of the acquired information is judged by a judgement part 24, and judgement result is displayed on a display part 25. Thereby, since the generation source of the acquired information by the radar can be classified automatically, it is possible to quickly deal with the acquired information effectively.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a categorization processing technique for determining what kind of object the source of information acquired by a Doppler radar is, and features extracted from the information acquired by the Doppler radar. The present invention relates to a target classification device that compares a characteristic with a characteristic of a classification target that already has and performs classification processing based on the matching ratio.

[0002]

2. Description of the Related Art As a conventional technique for classifying targets, for example, "target classifying device" (JP-A-4-22518).
The invention described in Japanese Patent No. 8) can be mentioned. This is because by detecting the target Doppler component using the active sonar signal to obtain velocity information, and removing the Doppler component of the passive sonar signal based on this velocity information,
It is an invention to obtain the fundamental frequency of a noise source.

Further, as a technique for obtaining the shape of an object, the invention described in "Device for measuring properties of moving body and method for measuring speed of moving body" (Japanese Patent Laid-Open No. 5-52950) can be mentioned. . This is an invention in which the shape, the moving speed, and the interval of the end portion of the moving body are obtained by using the Doppler radar.

[0004]

However, in the above-mentioned prior art, it is not possible to classify moving objects in detail. The purpose of the present invention is to analyze the information obtained by the Doppler radar by using the Doppler frequency for the comparison object, determine what the information was emitted from, and a target classification device that displays the determination result. To provide.

[0005]

According to the invention, the above mentioned objects are achieved by means of the patent claims.

That is, according to the first aspect of the present invention, as shown in FIG. 1, the radar section 10 for receiving the Doppler radar signal emitted from the relatively moving target and transmitting the Doppler radar information, and the Doppler radar information. An information analysis unit 11 that analyzes the information and outputs the analysis information.
A feature extraction unit 12 that extracts a feature from the analysis information and outputs the feature information, feature information based on a radar reception signal, and feature information that has been acquired in advance and represents a specific type of target object. And the feature matching unit 13 that outputs the matching result and the matching result are determined,
The target classification device includes a determination unit 14 that classifies a target object that is a generation source of a Doppler radar signal, and a display unit 15 that displays a classification result of the target.

The invention according to claim 2 is a target classifying device provided with an information analysis unit for performing high-speed Fourier transform processing to analyze Doppler radar information, and the invention according to claim 3 is
The target classifying device includes an information analyzing unit that performs wavelet analysis to analyze Doppler radar information. The invention of claim 4 is a target classifying device that includes an information analyzing unit that performs cepstrum processing to analyze Doppler radar information.

[0008]

According to the present invention, the characteristics of a moving object are extracted from the acquired information of the Doppler radar, and the characteristics are collated with the characteristics held in the characteristic table of each classification target that is already held. By determining the degree of matching, the operation is performed so as to determine what the acquired information was caused by.

[0009]

FIG. 2 shows a block diagram of an embodiment of the present invention. As shown in FIG. 2, the target classification device 26 according to the present embodiment includes a radar unit 20, an information analysis unit 21, a feature extraction unit 22, a feature matching unit 23, a determination unit 24, and a display unit 25. Hereinafter, each of the units 20 to 25 included in the target classification device 26 will be described.

The information acquired by the radar section 20 is sent to an information analysis section (FFT processing) 21, where information analysis by high-speed Fourier transform is performed. The analysis result is sent to the feature extraction unit 22. In the feature extraction unit 22,
The characteristics of the information acquired by the Doppler radar are changed to a form that can be compared with the characteristics of each classification target that has been acquired in advance. The output of the feature extraction unit 22 is sent to the feature matching unit 23.

The characteristic matching unit 23 compares the characteristics of the information acquired by the Doppler radar with the characteristics of each classification target that has been acquired in advance, and obtains the ratio of matching. And
The determination unit 24 determines what is the source of the acquired information, and the display unit 25 displays the determination result. According to this embodiment, the source of the information acquired by the radar can be specified automatically, so that the acquired information can be effectively dealt with quickly.

By the way, the information analysis unit 21 in the embodiment of FIG. 2 uses the FFT processing in order to analyze the information acquired by the Doppler radar, but a wavelet analysis method is also conceivable as another information analysis means.
The FFT processing is a method that emphasizes periodicity, while the wavelet analysis is an analysis method that emphasizes locality. FIG. 3 shows an embodiment of the present invention in which the wavelet analysis method is used.

Target classification device 3 in the embodiment shown in FIG.
6 is a radar unit 30, an information analysis unit 31, and a feature extraction unit 3
2, a feature matching unit 33, a determination unit 34, and a display unit 35. Among these, the radar unit 30 and the feature extraction unit 3
2, the feature matching unit 33, the determination unit 34, and the display unit 35 operate in the same manner as those having the same names shown in FIG.

On the other hand, the information analysis unit 31 differs from the information analysis unit 21 shown in FIG. 2 in that it performs wavelet analysis instead of FFT processing. Therefore, in the target classification device 36 shown in FIG. 3, particularly local features can be found.

A method using a cepstrum is also conceivable as an information analysis means other than FFT processing and wavelet analysis. The cepstrum is an analysis unit that performs FFT processing again on the information that has been FFT processed. In Figure 4,
An example of the present invention in which a cepstrum is used is shown.

Target classification device 4 in the embodiment shown in FIG.
6 is a radar unit 40, an information analysis unit 41, a feature extraction unit 4
2, a feature matching unit 43, a determination unit 44, and a display unit 45. Among these, the radar unit 40 and the feature extraction unit 4
2, the feature matching unit 43, the determining unit 44, and the display unit 45 operate in the same manner as those having the same names shown in FIG. 2 or FIG.

On the other hand, the information analysis unit 41 performs the cepstrum process instead of the FFT process or the wavelet analysis, and the information analysis units 21 and 31 shown in FIG. 2 or FIG.
Is different from. Therefore, in the target classifying device 46 shown in FIG.
It could be analyzed periodically rather than T-process.

FIG. 5 shows a flow chart for explaining the outline of processing in the apparatus of the present invention. The target classification process shown in the figure includes the steps of obtaining radar information, setting parameters required for FFT, etc., obtaining a spectrum or the like from which features are extracted, obtaining each feature, And the step of performing matching with the characteristics of.

In the flow chart of FIG. 5, when the characteristic matching process is completed for the specified number of data or the total number of data, the classification determination is performed and the targets are classified. And finally, the result is displayed. At the time of displaying the results, it is possible to display not only the classification result but also each feature of the spectrum data, the target speed, and the like.

[0020]

As described above, according to the present invention,
Since various information analyzes such as FFT, wavelet analysis, and cepstrum are performed, it is possible to accurately determine what caused the acquired information. Therefore, there is an advantage that effective countermeasures can be promptly taken against the source of the acquired information.

[Brief description of drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is a diagram showing an embodiment of the present invention.

FIG. 3 is a diagram showing another embodiment of the present invention.

FIG. 4 is a diagram showing another embodiment of the present invention.

FIG. 5 is a flowchart illustrating an outline of processing in the device of the present invention.

[Explanation of symbols]

 10, 20, 30, 40 Radar section 11, 21, 31, 41 Information analysis section 12, 22, 32, 42 Feature extraction section 13, 23, 33, 43 Feature matching section 14, 24, 34, 44 Judgment section 15, 25,35,45 Display unit 26,36,46 Target classification device

Claims (4)

[Claims] 1. A radar unit (10) for receiving a Doppler radar signal emitted from a relatively moving target and transmitting Doppler radar information, and for analyzing the Doppler radar information and outputting the analysis information. A part (11), a feature extraction part (12) for extracting a feature from the analysis information and outputting the feature information, a feature information based on a radar reception signal, and a feature information that has been acquired in advance and is of a specific type. A feature matching unit (1 that matches a target object and outputs the matching result.
3), a determination unit (14) that determines the matching result and classifies the target object that is the generation source of the Doppler radar signal, and a display unit (15) that displays the classification result of the target. Target classification device. 2. The information analysis unit for analyzing Doppler radar information by performing high-speed Fourier transform processing is provided.
The described target classification device. 3. The target classification device according to claim 1, further comprising the information analysis unit for performing wavelet analysis to analyze Doppler radar information. 4. The target classification device according to claim 1, further comprising the information analysis unit for performing cepstrum processing to analyze Doppler radar information.