CN106597400A - Ground moving vehicle target classification and recognition method and system based on high-resolution distance image - Google Patents

Abstract

The invention relates to a ground moving vehicle target classification and recognition method and system based on a high-resolution distance image, and the method comprises the steps: carrying out the averaging and energy normalization of multiple collected continuous original HRRP echoes of a ground moving vehicle target; calculating the HRRP echo power spectrum features according to the HRRP echoes after averaging and energy normalization; calculating the distances between the HRRP echo power spectrum features and various types of preset power spectrum feature templates, and obtaining a plurality of distances; comparing the distances, and determining the class of the ground moving vehicle target according to a comparison result. The method calculates the distances between the HRRP echo power spectrum features and various types of preset power spectrum feature templates through a nearest neighbor classifier, recognizes the class information of the ground moving vehicle target through comparing the distances between the HRRP echo power spectrum features and various types of preset power spectrum feature templates, achieves the precise classification and recognition of the ground moving vehicles, and provides support for the classification and recognition of military moving vehicles on the ground.

Description

Terrain vehicle target classification identification method and system based on High Range Resolution

Technical field

The present invention relates to vehicle recongnition technique field, and in particular to a kind of terrain vehicle based on High Range Resolution Target classification identification method and system.

Background technology

For terrain vehicle target, because different combat duties are needed with varying environment in modern battlefield environment Ask, combat duty that various vehicles are undertaken is different, also determine that its threat degree is different, thus to battlefield surroundings in it is various Vehicle carries out Classification and Identification to the ground reconnaissance in modern war, situation of battlefield perception, threat assessment, commanding and decision-making and accurately beats Hit with important value.

In low-resolution radar system, target vehicle only occupies one or a few range cell, and miscellaneous by ground Wave action is difficult the information for obtaining support target classification.Consider moving target in doppler spectral target Doppler and land clutter into Point separate, and target movable information included in doppler spectral, thus the doppler spectral of moving target provide for target classification can Energy.Due to wheeled vehicle different from creeper truck fine motion (vehicle wheel rotation of wheeled vehicle, the crawler belt rotation of creeper truck and putting down for upper crawler belt It is dynamic), existing result of study shows, the doppler spectral of terrain vehicle has separability, using the doppler spectral of moving vehicle Feature can will be divided into wheeled vehicle and the big class of creeper truck two to terrain vehicle.

Due to the restriction of resolution, its rough segmentation can only be wheeled vehicle and creeper truck to ground moving object by low-resolution radar Two big class, will realize that the concrete identification to target model is more difficult.

The content of the invention

The technical problem to be solved is to provide a kind of terrain vehicle target based on High Range Resolution Classifying identification method and system.

The technical scheme that the present invention solves above-mentioned technical problem is as follows：A kind of ground motion car based on High Range Resolution Target classification identification method, comprises the following steps：

S1, the multiple continuous original high-resolution Range Profile HRRP echoes of the terrain vehicle target to gathering are carried out averagely And energy normalized, obtain the HRRP echoes after average and energy normalized；

S2, according to the HRRP echoes after average and energy normalized, calculates the HRRP echo work(of terrain vehicle target Rate spectrum signature；

S3, calculates the HRRP echo power spectrum characteristics and default various power spectrum characteristic templates of terrain vehicle target The distance between, obtain multiple distances；

S4, the plurality of distance is compared, and according to comparative result the class of vehicle of terrain vehicle target is determined.

The invention has the beneficial effects as follows：It is various with default HRRP echo power spectrum characteristics to be calculated using nearest neighbor classifier The distance between power spectrum characteristic template, by comparing HRRP echo power spectrum characteristics with default multiple power spectrum signature template The distance between recognizing the classification information of terrain vehicle target, realize the precise classification to terrain vehicle and know Not, the Classification and Identification for the military moving vehicle on ground provides support.

On the basis of above-mentioned technical proposal, the present invention can also do following improvement.

Further, the S1 is specifically included：

S11, the multiple continuous original high-resolution Range Profile HRRP echoes of the terrain vehicle target to gathering carry out miscellaneous Ripple suppresses, and obtains the multiple continuous HRRP echoes after clutter recognition；

S12, the multiple continuous HRRP echoes after clutter recognition carry out average and energy normalized.

It is using the beneficial effect of above-mentioned further scheme：HRRP echoes to gathering carry out clutter recognition, improve The signal to noise ratio of HRRP echoes, reduces the impact that clutter is recognized to final terrain vehicle.

Further, the S11 is specifically included：

S11.1, the multiple continuous HRRP echoes of collection terrain vehicle target, are S=[s₁..., s_i..., s_N], i =1 ..., N, wherein, s_iRepresent the i ＆ lt HRRP echo of terrain vehicle target；

S11.2, carries out Fourier transform and obtains target range-doppler image R to multiple continuous HRRP echoes S；

S11.3, by the passage zero setting that Doppler frequency in the target range-doppler image R is 0, after obtaining zero setting Target range-doppler image R '；

S11.4, carries out inverse Fourier transform, after obtaining clutter recognition to the target range after zero setting-doppler image R ' HPPR echo S '.

Further, the S12 is specifically included：

S12.1, according to the HRRP echo S ' after clutter recognition, calculates target average distance pictureIts In, s '_iFor S ' i-th arranges, and abs () is the operation that takes absolute value, s_mFor target average distance picture；

S12.2, to target average distance as s_mEnergy normalized is carried out, the HRRP echoes after energy normalized are obtainedWherein norm () is represented and is asked vectorial two norm to operate.

Further, the S2 calculates the HRRP echo power spectrum characteristics of terrain vehicle target by equation below：

F=FFT (s '_m)

X (i)=F (i) F^H(i), i=1,2 ..., ceil (M/2)

Wherein ()^HConjugate operation is represented, ceil () represents the operation that rounds up, and F is target normalization average distance As s '_mFourier transformation, M for F length, x (i) for target power spectrum signature x i-th element.

Further, the S3 is specifically included：

Using HRRP echo power spectrum characteristics x (i) of terrain vehicle target as nearest neighbor classifier input, according to Following formula by nearest neighbor classifier calculate between HRRP echo power spectrum characteristics and default all kinds of power spectral templates away from From：

Wherein, min () represents minimum operation,For j-th template of the i-th class power spectrum characteristic, L_iRepresent the i-th class Power spectrum characteristic template number, C represents power spectrum characteristic classification number, d_iRepresent power spectrum characteristic x and the i-th class power spectrum characteristic Distance.

Further, the S4 is specifically included：

By car corresponding with the power spectral templates of the HRRP echo power spectrum characteristics of terrain vehicle target distance minimum The class of vehicle of classification as terrain vehicle target.

To solve the technical problem of the present invention, a kind of terrain vehicle target based on High Range Resolution is additionally provided Classifying and identifying system, including：

Average normalized module, for continuous original high-resolution Range Profile multiple to the terrain vehicle target for gathering HRRP echoes carry out average and energy normalized, obtain the HRRP echoes after average and energy normalized；

Spectra calculation module, for according to the HRRP echoes after average and energy normalized, calculating terrain vehicle The HRRP echo power spectrum characteristics of target；

Distance calculation module, the HRRP echo power spectrum characteristics for calculating terrain vehicle target are various with default The distance between power spectrum characteristic template, obtains multiple distances；

Class of vehicle identification module, for the plurality of distance to be compared, according to comparative result ground motion is determined The class of vehicle of vehicle target.

Further, the average normalized module also includes：

Clutter recognition module, for continuous original high-resolution Range Profile multiple to the terrain vehicle target for gathering HRRP echoes carry out clutter recognition, obtain the multiple continuous HRRP echoes after clutter recognition；

The average normalized module to clutter recognition after multiple continuous HRRP echoes carry out average and energy normalizing Change.

Further, the class of vehicle identification module specifically for：

By car corresponding with the power spectral templates of the HRRP echo power spectrum characteristics of terrain vehicle target distance minimum The class of vehicle of classification as terrain vehicle target.

Description of the drawings

Fig. 1 is a kind of terrain vehicle target identification method flow chart based on High Range Resolution of embodiment 1；

Fig. 2 is a kind of terrain vehicle target identification method flow chart based on High Range Resolution of embodiment 2；

Fig. 3 is a kind of schematic frame of terrain vehicle target identification system based on High Range Resolution of embodiment 3 Figure；

Fig. 4 is a kind of schematic frame of terrain vehicle target identification system based on High Range Resolution of embodiment 4 Figure.

Specific embodiment

The principle and feature of the present invention are described below in conjunction with accompanying drawing, example is served only for explaining the present invention, and It is non-for limiting the scope of the present invention.

As shown in figure 1, for a kind of terrain vehicle target identification method based on High Range Resolution of embodiment 1, Including：

S1, the multiple continuous original high-resolution Range Profile HRRP echoes of the terrain vehicle target to gathering are carried out averagely And energy normalized, obtain the HRRP echoes after average and energy normalized；

S2, according to the HRRP echoes after average and energy normalized, calculates the HRRP echo work(of terrain vehicle target Rate spectrum signature；

S3, calculates the HRRP echo power spectrum characteristics and default various power spectrum characteristic templates of terrain vehicle target The distance between, obtain multiple distances；

S4, the plurality of distance is compared, and according to comparative result the class of vehicle of terrain vehicle target is determined.

It should be understood that first to the vehicle target in ground motion, such as, the vehicle moved on military battlefield, collection should The multiple continuous High Range Resolution HRRP echo of vehicle target, and average and energy normalized is carried out, obtain average and energy HRRP echoes after normalization.The multiple continuous HRRP echoes of the present embodiment collection vehicle target, then carry out average and normalizing Change, the HRRP echoes after average and normalization can more be reflected in the moving situation of the vehicle target of ground motion.Then, to flat HRRP echoes and after normalization, calculate the HRRP echo power spectrum characteristics of terrain vehicle target, by power spectrum characteristic The distance between with default various power spectrum characteristic templates, by the HRRP echo power spectrum characteristics of terrain vehicle target The distance between with default various power spectrum characteristic templates, can Classification and Identification go out the class of vehicle of moving vehicle target, be The Classification and Identification of the military moving vehicle on ground is provided and supported.

As shown in Fig. 2 for a kind of terrain vehicle target identification method based on High Range Resolution of embodiment 2, Including：

S1 ', the multiple continuous original high-resolution Range Profile HRRP echoes of the terrain vehicle target to gathering carry out miscellaneous Ripple suppression is processed；

S2 ', to the multiple continuous HRRP echoes after clutter recognition process average and energy normalized is carried out, and obtains average With the HRRP echoes after energy normalized；

S3 ', according to the HRRP echoes after average and energy normalized, calculates the HRRP echo work(of terrain vehicle target Rate spectrum signature；

S4 ', calculates the HRRP echo power spectrum characteristics and default various power spectrum characteristic moulds of terrain vehicle target The distance between plate, obtains multiple distances；

S5 ', the plurality of distance is compared, and according to comparative result the vehicle class of terrain vehicle target is determined Not.

It should be understood that the method for the present embodiment is entered in the multiple continuous HRRP echoes of the terrain vehicle target to gathering Row carries out clutter recognition, it is possible to increase the letter of HRRP echoes averagely and before energy normalized to multiple continuous HRRP echoes Miscellaneous ratio, reduces impact of the clutter to result after average and energy normalized, and then finally can more accurately to moving vehicle target Carry out Classification and Identification.

Wherein, the multiple continuous original high-resolution Range Profile HRRP echoes of terrain vehicle target to gathering carry out miscellaneous Ripple suppress process detailed process be：

The multiple continuous HRRP echoes of collection terrain vehicle target, are S=[s₁..., s_i..., s_N], i= 1 ..., N, wherein, s_iRepresent the i ＆ lt HRRP echo of terrain vehicle target；

Fourier transform is carried out to multiple continuous HRRP echoes S and obtains target range-doppler image R, wherein, R= FFT1 (S), FFT1 () represent that the every a line to matrix does Fourier transformation；

By the passage zero setting that Doppler frequency in the target range-doppler image R is 0, the target after zero setting is obtained Distance-Doppler image R '；

Inverse Fourier transform is carried out to the target range after zero setting-doppler image R ', the HPPR after clutter recognition is obtained Echo S ', S '=IFFT1 (R '), IFFT1 () represent that the every a line to matrix does inverse Fourier transform.

HRRP echo S ' after clutter recognition are obtained by said method, then the HRRP echo S ' after clutter recognition are entered Row is average and energy normalized, and detailed process is：

According to the HRRP echo S ' after clutter recognition, target average distance picture is calculatedWherein, s′_iFor S ' i-th arranges, and abs () is the operation that takes absolute value, s_mFor target average distance picture；

To target average distance as s_mEnergy normalized is carried out, the HRRP echoes after energy normalized are obtainedWherein norm () is represented and is asked vectorial two norm to operate.

The multiple continuous HRRP echoes of the terrain vehicle target to gathering carry out average and energy normalized and process, HRRP echoes after average and normalization can more be reflected in the moving situation of the vehicle target of ground motion.

The HRRP echoes of the terrain vehicle target to gathering are carried out after the process of average and energy normalized, calculate HRRP Echo power spectrum characteristics, specifically by equation below the HRRP echo power spectrum characteristics of terrain vehicle target are calculated：

Wherein ()^HConjugate operation is represented, ceil () represents the operation that rounds up, and F is target normalization average distance As s '_mFourier transformation, M for F length, x (i) for HRRP echo power spectrum characteristics x i-th element.

Subsequently, the HRRP echo power spectrum characteristics and default all kinds of power spectrum characteristics of terrain vehicle target will be calculated The distance between module, specially：

Using HRRP echo power spectrum characteristics x (i) of terrain vehicle target as nearest neighbor classifier input, according to Following formula by nearest neighbor classifier calculate between HRRP echo power spectrum characteristics and default all kinds of power spectral templates away from From：

Wherein, min () represents minimum operation,For j-th template of the i-th class power spectrum characteristic, L_iRepresent the i-th class Power spectrum characteristic template number, C represents power spectrum characteristic classification number, d_iRepresent power spectrum characteristic x and the i-th class power spectrum characteristic Distance.

It will be appreciated that default power spectrum characteristic has plurality of classes, the multiple moulds of power spectrum characteristic correspondence of each classification Plate, for preset each classification power spectrum characteristic, calculate terrain vehicle target HRRP echo power spectrum characteristics with it is pre- If the distance between the various template of each classification power spectrum characteristic, using minimum range as terrain vehicle target HRRP The distance between power spectrum characteristic template of echo power spectrum characteristics and the pre-set categories.Terrain vehicle target HRRP echo The distance between power spectrum characteristic and power spectrum characteristic template of each pre-set categories are calculated, by terrain vehicle The distance between target HRRP echo power spectrum characteristics and power spectrum characteristic template of each pre-set categories are compared, will be with ground The moving vehicle target HRRP echo power spectrum characteristics distance minimum corresponding classification of power spectrum characteristic template in face is transported as ground The class of vehicle of dynamic vehicle target.HRRP echo power spectrum characteristics are calculated using nearest neighbor classifier to compose with default multiple power The distance between feature templates, by comparing between HRRP echo power spectrum characteristics and default multiple power spectrum signature template Distance realizes the precise classification to terrain vehicle and recognizes recognizing the classification information of terrain vehicle target, is ground The Classification and Identification of the military moving vehicle on face is provided and supported.

The method that the application is provided is further detailed below by measured data：

1st, experiment scene：

Measurement data of the measured data comprising three kinds of vehicle targets, respectively IVECO, truck, Vehicle De L'Avant Blinde By Creussot, three kinds Vehicle carries out circular motion with the circle that radius is 50 meters respectively, the HRRP during being turn-taked for the first time to three kinds of targets with 0.5 degree Echo carries out framing, and the HRRP echoes per frame in are after clutter recognition, average, energy normalized, feature extraction as target Template base.Test sample is the HRRP data that three kinds of vehicle targets are turn-taked for the second time.

2., experiment content：

2.1) for certain original HRRP signal carries out clutter recognition, the clutter of Vehicle De L'Avant Blinde By Creussot, truck, IVECO is obtained HRRP echoes after suppression；

2.2) HRRP after clutter recognition is carried out averagely, and energy normalized is carried out to it；

2.3) target power spectrum signature is calculated according to the average HRRP of energy normalized；

2.4) the distance between target power spectrum signature and all kinds of power spectral templates are calculated, according to obtaining apart from comparative result Class of vehicle information, based on measured data, the classification results of the application are as shown in table 1：

Table 1

	Truck	IVECO	Vehicle De L'Avant Blinde By Creussot
				Truck	0.9780	0.0066	0.0153
IVECO	0.0110	0.8616	0.1274
				Vehicle De L'Avant Blinde By Creussot	0.0219	0.1420	0.8361

In table 1, the true classification of row representative sample, row represent the recognition result class after being identified by the application method Not, data therein represent the accuracy rate of identification, the recognition methodss for being provided by the application as can be seen from Table 1, identification Accurate rate is that comparison is high, is substantially higher than 80%.

As shown in figure 3, a kind of terrain vehicle target classification identification based on High Range Resolution for embodiment 3 is System, including average normalized module 31, spectra calculation module 32, distance calculation module 33 and vehicle difference identification module 34.

Average normalized module 31, for continuous original high-resolution distance multiple to the terrain vehicle target for gathering As HRRP echoes carry out average and energy normalized, the HRRP echoes after average and energy normalized are obtained.

Spectra calculation module 32, for the HRRP according to average normalized module 31 averagely and after energy normalized process Echo, calculates the HRRP echo power spectrum characteristics of terrain vehicle target,

Distance calculation module 33, for calculating the terrain vehicle target calculated according to spectra calculation module 32 The distance between HRRP echo power spectrum characteristics and default various power spectrum characteristic templates, obtain multiple distances.

Class of vehicle identification module 34, the multiple distances for distance calculation module 33 to be calculated are compared, according to Comparative result determines the class of vehicle of terrain vehicle target.

The class of vehicle identification module 34 specifically for：Will be special with the HRRP echo powers of terrain vehicle target spectrum The minimum corresponding class of vehicle of power spectral templates of distance is levied as the class of vehicle of terrain vehicle target.

Mould is recognized using average normalized module 31, spectra calculation module 32, distance calculation module 33 and vehicle classification Block 34 realizes that the technical characteristic of the Classification and Identification of terrain vehicle target may be referred to the correlation technique feature in embodiment, then It is secondary to repeat no more.

As shown in figure 4, a kind of terrain vehicle target classification identification based on High Range Resolution for embodiment 4 is System, including clutter recognition module 41, average normalized module 42, spectra calculation module 43, distance calculation module 44 and vehicle Difference identification module 45.

Clutter recognition module 41, for continuous original high-resolution Range Profile multiple to the terrain vehicle target for gathering HRRP echoes carry out clutter recognition, obtain the multiple continuous HRRP echoes after clutter recognition.

Average normalized module 42, it is multiple continuous after clutter recognition process for carrying out to clutter recognition module 41 HRRP echoes carry out average and energy normalized, obtain the HRRP echoes after average and energy normalized.

Spectra calculation module 43, for the HRRP according to average normalized module 42 averagely and after energy normalized process Echo, calculates the HRRP echo power spectrum characteristics of terrain vehicle target,

Distance calculation module 44, for calculating the terrain vehicle target calculated according to spectra calculation module 43 The distance between HRRP echo power spectrum characteristics and default various power spectrum characteristic templates, obtain multiple distances.

Class of vehicle identification module 45, the multiple distances for distance calculation module 44 to be calculated are compared, according to Comparative result determines the class of vehicle of terrain vehicle target.

A kind of terrain vehicle target classification identification method and system based on High Range Resolution that the present invention is provided, HRRP echoes to gathering carry out clutter recognition, improve the signal to noise ratio of HRRP echoes, reduce clutter to final ground sport(s) car Identification impact；HRRP echo power spectrum characteristics and default multiple power spectrum signature template are calculated using nearest neighbor classifier The distance between, known by comparing the distance between HRRP echo power spectrum characteristics and default multiple power spectrum signature template The classification information of other terrain vehicle target, realizes the precise classification to terrain vehicle and recognizes, is the army on ground The Classification and Identification of thing moving vehicle is provided and supported.

The foregoing is only presently preferred embodiments of the present invention, not to limit the present invention, all spirit in the present invention and Within principle, any modification, equivalent substitution and improvements made etc. should be included within the scope of the present invention.

Claims (10)

1. a kind of terrain vehicle target classification identification method based on High Range Resolution, it is characterised in that including following Step：

S1, the multiple continuous original high-resolution Range Profile HRRP echoes of the terrain vehicle target to gathering carry out average and energy Amount normalization, obtains the HRRP echoes after average and energy normalized；

S2, according to the HRRP echoes after average and energy normalized, calculates the HRRP echo powers spectrum of terrain vehicle target Feature；

S3, calculates between the HRRP echo power spectrum characteristics and default various power spectrum characteristic templates of terrain vehicle target Distance, obtain multiple distances；

S4, the plurality of distance is compared, and according to comparative result the class of vehicle of terrain vehicle target is determined.

2. the terrain vehicle target classification identification method of High Range Resolution, its feature are based on as claimed in claim 1 It is that the S1 is specifically included：

S11, the multiple continuous original high-resolution Range Profile HRRP echoes of the terrain vehicle target to gathering carry out clutter suppression System, obtains the multiple continuous HRRP echoes after clutter recognition；

S12, the multiple continuous HRRP echoes after clutter recognition carry out average and energy normalized.

3. the terrain vehicle target classification identification method of High Range Resolution, its feature are based on as claimed in claim 2 It is that the S11 is specifically included：

S11.1, the multiple continuous HRRP echoes of collection terrain vehicle target, are S=[s₁..., s_i..., s_N], i= 1 ..., N, wherein, s_iRepresent the i ＆ lt HRRP echo of terrain vehicle target；

S11.2, carries out Fourier transform and obtains target range-doppler image R to multiple continuous HRRP echoes S；

S11.3, by the passage zero setting that Doppler frequency in the target range-doppler image R is 0, obtains the mesh after zero setting Subject distance-doppler image R '；

S11.4, carries out inverse Fourier transform, after obtaining clutter recognition to the target range after zero setting-doppler image R ' HPPR echo S '.

4. the terrain vehicle target classification identification method of High Range Resolution, its feature are based on as claimed in claim 3 It is that the S12 is specifically included：

S12.1, according to the HRRP echo S ' after clutter recognition, calculates target average distance pictureIts In, s '_iFor S ' i-th arranges, and abs () is the operation that takes absolute value, s_mFor target average distance picture；

S12.2, to target average distance as s_mEnergy normalized is carried out, the HRRP echoes after energy normalized are obtainedWherein norm () is represented and is asked vectorial two norm to operate.

5. the terrain vehicle target classification identification method of High Range Resolution, its feature are based on as claimed in claim 4 It is that the S2 calculates the HRRP echo power spectrum characteristics of terrain vehicle target by equation below：

$\begin{array}{c}F=FFT\left(s_m^{\′}\right)\\ x\left(i\right)=F\left(i\right)\·F^H\left(i\right),i=1,2,...,ceil(M/2)\end{array};$

Wherein ()^HConjugate operation is represented, ceil () represents the operation that rounds up, and F is target normalization average distance as s '_m Fourier transformation, M for F length, x (i) for target power spectrum signature x i-th element.

6. the terrain vehicle target classification identification method of High Range Resolution, its feature are based on as claimed in claim 5 It is that the S3 is specifically included：

Using HRRP echo power spectrum characteristics x (i) of terrain vehicle target as nearest neighbor classifier input, according to following Formula calculates the distance between HRRP echo power spectrum characteristics and default all kinds of power spectral templates by nearest neighbor classifier：

$d_i=\underset{j}{min}(x-x_i^j),j=1,2,...,L_i;i=1,2,...,C;$

Wherein, min () represents minimum operation,For j-th template of the i-th class power spectrum characteristic, L_iRepresent the i-th class power Spectrum signature template number, C represents power spectrum characteristic classification number, d_iRepresent power spectrum characteristic x and the i-th class power spectrum characteristic away from From.

7. the terrain vehicle target classification identification method of High Range Resolution, its feature are based on as claimed in claim 6 It is that the S4 is specifically included：

By vehicle class corresponding with the power spectral templates of the HRRP echo power spectrum characteristics of terrain vehicle target distance minimum Not as the class of vehicle of terrain vehicle target.

8. a kind of terrain vehicle target classification identifying system based on High Range Resolution, it is characterised in that include：

Average normalized module, for continuous original high-resolution Range Profile HRRP multiple to the terrain vehicle target for gathering Echo carries out average and energy normalized, obtains the HRRP echoes after average and energy normalized；

Spectra calculation module, for according to the HRRP echoes after average and energy normalized, calculating terrain vehicle target HRRP echo power spectrum characteristics；

Distance calculation module, for calculating the HRRP echo power spectrum characteristics and default various power of terrain vehicle target The distance between spectrum signature template, obtains multiple distances；

Class of vehicle identification module, for the plurality of distance to be compared, according to comparative result terrain vehicle is determined The class of vehicle of target.

9. the terrain vehicle target classification identifying system of High Range Resolution, its feature are based on as claimed in claim 8 It is that the average normalized module also includes：

Clutter recognition module, for repeatedly continuous original high-resolution Range Profile HRRP to be returned to the terrain vehicle target for gathering Ripple carries out clutter recognition, obtains the multiple continuous HRRP echoes after clutter recognition；

The average normalized module to clutter recognition after multiple continuous HRRP echoes carry out average and energy normalized.

10. the terrain vehicle target classification identifying system of High Range Resolution, its feature are based on as claimed in claim 9 Be, the class of vehicle identification module specifically for：

By vehicle class corresponding with the power spectral templates of the HRRP echo power spectrum characteristics of terrain vehicle target distance minimum Not as the class of vehicle of terrain vehicle target.