CN102841341A - Pulse radar moving-target detection method - Google Patents

Abstract

The invention relates to a pulse radar moving-target detection method comprising the following steps of: setting radar system parameters to obtain the size of a clutter map; calculating a threshold value according to the radar system parameters; performing two-dimension DCT (discrete cosine transform) on the clutter map; filtering matrix data which is subjected to the two-dimension DCT according to a threshold value; performing IDCT (inverse discrete cosine transform) on the filtered matrix data, and reducing the clutter map; and detecting the reduced clutter map. By performing the two-dimension DCT on the clutter map in the pulse radar moving-target detection method, the DCT calculating quantity is low, and the detection method is clear in physical significance and high in precision; meanwhile, the threshold value filtering is used without aiming at special frequency, so that the method is excellent in adaptability. With the adoption of the pulse radar moving-target detection method in which the two-dimension DCT is adopted, the calculation process can be simplified, and meanwhile, the higher adaptability is achieved.

Description

A kind of pulsed radar moving target detection method

Technical field

The present invention relates to the pulsed radar moving target detection technique, relate in particular to a kind of pulsed radar moving target detection method based on two-dimensional dct (discrete cosine transform) filtering.

Background technology

The ultra-wideband pulse radar has been widely used in the various aspects of social production, and positioning performance that it is superior and eurytopicity can receive various circles of society's favorable comment deeply.Yet radar wave receives the influence of various factors in transmission course; The echo that receives has also been swept along very many undesired signal and noises except that the motor message that carries target.The existence of these interference and noise has had a strong impact on the detection probability and the CFAR performance of radar.

Usually algorithm only carries out the filtering of one dimension to slow time shaft, does not use the quantity of information of fast time shaft, and effect is undesirable.Adopt traditional FIR (finite impulse response) and the algorithm of FFT (Fast Fourier Transform (FFT)), operand is big, in addition; Through windowing iterative computation two-dimensional filtering, itself be exactly a thing that consumes very much calculation resources, simultaneously because the windowing truncated data; To cause time-domain signal by broadening, introduce distortion, especially guarantee under the situation of certain investigative range at needs; Number of data points is increased sharply, and operand further strengthens.At present, though the Radar Signal Processing method based on DCT occurred, be the one dimension dct transform, so only utilized the quantity of information of one-dimensional signal, degree of accuracy is low and operand is big.

Summary of the invention

Based on this, be necessary defective to above-mentioned pulsed radar moving target detection technique existence, provide a kind of degree of accuracy the high little a kind of pulsed radar moving target detection method of operand simultaneously.

A kind of pulsed radar moving target detection method comprises: the radar system parameter is set to obtain the clutter map size; According to said radar system calculation of parameter threshold value; Said clutter map is carried out two-dimensional dct transform; According to said threshold value the matrix data through two-dimensional dct transform is carried out filtering; To carrying out idct transform through filtered matrix data, the reduction clutter map; And the clutter map after the said reduction is detected.

In the present embodiment, wherein, the radar system parameter is set, comprises the steps: according to radar performance and sensing range selected frame size to obtain clutter map; According to motion of objects characteristic Design sampling system to be detected; Reach sampling precision, obtain the clutter map size according to said sampling system needs.

In the present embodiment, said threshold value comprises: vertical low frequency threshold value Tl, vertical high frequency threshold value Th, horizontal low frequency threshold value Tsl and horizontal high frequency threshold value Tsh; Wherein, the computing method of said vertical low frequency threshold value Tl are: Tl=f _Low* M/ (f _V/ 2)

In the formula, f _LowBe object motion frequency to be detected, M is the line number of said clutter map, f _VBe slow time frequency acquisition; Wherein, the computing method of said vertical high frequency threshold value Th are:

Th＝f _hi*M/(f _V/2)

In the formula, f _HiBe object motion frequency to be detected, M is the line number of said clutter map, f _VBe slow time frequency acquisition; Wherein, the computing method of said horizontal low frequency threshold value Tsl are:

Tsl＝f _sl*N/(f _H/2)

In the formula, f _SlBe real pulse shape, N is the columns of said clutter map, f _HBe fast time frequency acquisition;

Wherein, the computing method of said horizontal high frequency threshold value Tsh are:

Tsh＝f _sh*N/(f _H/2)

In the formula, f _ShBe real pulse shape, N is the columns of said clutter map, f _HBe fast time frequency acquisition.

In the present embodiment, wherein, said clutter map is carried out two-dimensional dct transform realizes according to the configurations formula:

$F(p,q)=a_p{a}_q\underset{m=0}{\overset{M-1}{\mathrm{\Σ}}}\underset{n=0}{\overset{N-1}{\mathrm{\Σ}}}{A}_{\mathrm{mn}}\cos \frac{\mathrm{\π}(2m+1)p}{2M}\cos \frac{\mathrm{\π}(2n+1)q}{2N}$

Wherein, in the formula, the ordinate of p remarked pixel, the horizontal ordinate of q remarked pixel

$a_p=\left\{\begin{array}{cc}\frac{1}{\sqrt{M}},& p=0\\ \sqrt{\frac{2}{M}},& 1\≤p\≤M-1\end{array}\right.,$ $a_q=\left\{\begin{array}{cc}\frac{1}{\sqrt{N}},& q=0\\ \sqrt{\frac{2}{N}},& 1\≤q\≤N-1\end{array}\right.,$

0≤p≤M-1,0≤q≤N-1, a _p, a _qBe one group of coefficient of dct transform, m representes the horizontal ordinate of original image pixel, and n representes the ordinate of original image pixel; M representes the quantity of original image pixels across point; N representes the quantity of the vertical pixel of original image, and Amn representes that coordinate is (m, the value of pixel n) in the original image.

In the present embodiment, wherein, according to said threshold value the matrix data through two-dimensional dct transform is carried out filtering and realizes according to the configurations formula:

$\left\{\begin{array}{ccc}B(i,j)=0& i<\mathrm{Tlori}>\mathrm{Th},& j<\mathrm{Tslorj}>\mathrm{Tsh}\\ B(i,j)=B(i,j)& \mathrm{Tl}<i<\mathrm{Th},& \mathrm{Tsl}<j<\mathrm{Tsh}\end{array}\right.$

In the formula, (i, j) coordinate is (i, the value of pixel j) to B in the transform domain image behind the expression two-dimensional dct transform.

In the present embodiment, wherein, to carrying out idct transform through filtered matrix data, the reduction clutter map is realized according to the configurations formula:

$A_{\mathrm{mn}}=\underset{m=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{a}_p{a}_{q}F(p,q)\cos \frac{\mathrm{\&pi;}(2m+1)p}{2M}\cos \frac{\mathrm{\&pi;}(2n+1)q}{2N}$

Wherein, in the formula, p representes the ordinate of transform domain pixel, and q representes the horizontal ordinate of transform domain pixel

$a_p=\left\{\begin{array}{cc}\frac{1}{\sqrt{M}},& p=0\\ \sqrt{\frac{2}{M}},& 1\&le;p\&le;M-1\end{array}\right.,$ $a_q=\left\{\begin{array}{cc}\frac{1}{\sqrt{N}},& q=0\\ \sqrt{\frac{2}{N}},& 1\&le;q\&le;N-1\end{array}\right.,$

0≤p≤M-1,0≤q≤N-1, a _pA _qIt is one group of coefficient of dct transform.M representes the horizontal ordinate of original image pixel, and n representes the ordinate of original image pixel, and M representes the quantity of original image pixels across point, and N representes the quantity of the vertical pixel of original image, and Amn representes that coordinate is (m, the value of pixel n) in the original image.

In the present embodiment, wherein, the clutter map after the said reduction is detected the target function value that comprises the steps: to calculate each point that lists on the clutter map after the reduction; Record surpasses the corresponding pixel position of target function value of amplitude threshold; Add up above-mentioned pixel position, and calculate its probability density; And select the big point of all density, be the target location.

In the present embodiment, said target function value is a variance yields, and the computing method of said variance yields are:

$D=\underset{t=1}{\overset{N}{\mathrm{\&Sigma;}}}{[x_i\left(t\right)-\mathrm{mean}\left(x\right)]}^2$

In the formula, x _i(t) value of the file pixel on the back image of i.e. reduction on the expression respective distances picture, mean (x) representes the average of these all samples of file pixel.

Said target function value is the kurtosis value of each row, and the computing method of said kurtosis value are:

$K_4^i\left(t\right)=K_4^i(t-1)-\mathrm{\&mu;T}(K_4^i(t-1)-x_i{(t-1)}^4+3)$

Wherein: T is the sampling period, and μ gets 60*T, x usually _i(t) value of the file pixel on the back image of i.e. reduction on the expression respective distances picture,

The expression kurtosis.

Above-mentioned a kind of pulsed radar moving target detection method is through carrying out two-dimensional dct transform to clutter map, and the dct transform calculated amount is less, and has clear physical meaning, and degree of accuracy is high; The threshold filter that adopts simultaneously to CF, does not have better adaptability.Can simplify calculating through the pulsed radar moving target detection method that adopts two-dimensional dct filtering and have better adaptability simultaneously.

Description of drawings

The schematic flow sheet of a kind of pulsed radar moving target detection method that Fig. 1 provides for the embodiment of the invention.

Fig. 2 is provided with the radar system parameter to obtain the schematic flow sheet of clutter map for what the embodiment of the invention provided.

The schematic flow sheet that Fig. 3 provides for the embodiment of the invention to the detection of the clutter map after the reduction.

Fig. 4 goes the average processing to obtain new clutter map for what present embodiment provided to each frame.

Fig. 5 carries out the clutter map B that two-dimensional dct transform obtains for provided by the invention to clutter map.

Fig. 6 for the embodiment of the invention provide to carry out the clutter map C that idct transform obtains through filtered matrix data.

The pattern that 4 windows are carried out Conjoint Analysis that Fig. 7 provides for the embodiment of the invention.

The final decision diagram that Fig. 8 provides for the embodiment of the invention.

Embodiment

See also Fig. 1, the schematic flow sheet of a kind of pulsed radar moving target detection method that provides for the embodiment of the invention comprises the steps:

Step S10: the radar system parameter is set to obtain the clutter map size.

See also Fig. 2, the radar system parameter is set to obtain the schematic flow sheet of clutter map for what the embodiment of the invention provided, step S10 is specially:

Step S11: according to radar performance and sensing range selected frame size.

Radar performance refers to that radar itself is to signals sampling speed with to the storage capacity of information.The minor increment resolution of supposing radar is R, and storage depth is M, and then the distance of frame data representative is R*M; If R and M are adjustable, then adjust the value of R and M as required, make it satisfy real needs.

Step S12: according to motion of objects characteristic Design sampling system to be detected.

Sampling system should design according to practical application.For the application of appointment, if the slow time maximum frequency deviation that causes of known motion of objects characteristic to be detected is f, 3～5 times that then choose SF and be f get final product, and need not to pursue accurate sampling radar mean frequency signal; And for the performance decision of fast time-sampling frequency by radar itself, corresponding with the fastest time that radar hardware can be made a response.

Step S13:, obtain the clutter map size according to the sampling precision of sampling system needs.

Obtain slow time-sampling frequency according to step S12, and fast time-sampling precision is determined by practical application, takes all factors into consideration range resolution R, the frame number num-M that representes on the fast time shaft, the parameters such as frequency resolution Rm on the slow time shaft.If slow temporal frame per second is nr, then in order to reach the frequency resolution Rm on the slow time shaft, the one-tenth frame number N=nr/Rm that needs; Then clutter map is of a size of: M * num-M * N, and wherein M is counting of fast time one frame, N is the image longitudinal size.

Step S20: according to radar system calculation of parameter threshold value.

The radar system parameter can comprise: fast time-sampling frequency is f _H, slow time-sampling frequency is f _VAnd the clutter map size of intercepting, be designated as the capable N row of M; Threshold value comprises: vertical low frequency threshold value Tl, vertical high frequency threshold value Th, horizontal low frequency threshold value Tsl and horizontal high frequency threshold value Tsh; Here getting object motion frequency to be detected is that 0.2Hz～5Hz also is designated as f respectively _Low～f _Hi, real pulse shape f _Sl～f _Sh, then:

Tl＝f _low*M/(f _V/2)

Th＝f _hi*M/(f _V/2)

Tsl＝f _sl*N/(f _H/2)

Tsh＝f _sh*N/(f _H/2)。

Step S30: clutter map is carried out two-dimensional dct transform.

DCT is a kind of real number field conversion, and its transformation kernel is the real number cosine function; After piece image carried out discrete cosine transform, the important informations of many relevant images all concentrated in the sub-fraction coefficient of dct transform.In embodiment provided by the invention, clutter map is carried out two-dimensional dct transform realizes according to the configurations formula:

$F(p,q)=a_p{a}_q\underset{m=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{A}_{\mathrm{mn}}\cos \frac{\mathrm{\&pi;}(2m+1)p}{2M}\cos \frac{\mathrm{\&pi;}(2n+1)q}{2N}$

Wherein, in the formula, p representes the ordinate of transform domain pixel, and q representes the horizontal ordinate of transform domain pixel

$a_p=\left\{\begin{array}{cc}\frac{1}{\sqrt{M}},& p=0\\ \sqrt{\frac{2}{M}},& 1\&le;p\&le;M-1\end{array}\right.,$ $a_q=\left\{\begin{array}{cc}\frac{1}{\sqrt{N}},& q=0\\ \sqrt{\frac{2}{N}},& 1\&le;q\&le;N-1\end{array}\right.,$

0≤p≤M-1，0≤q≤N-1，

a _p, a _qBe one group of coefficient of dct transform, m representes the horizontal ordinate of original image pixel, and n representes the ordinate of original image pixel; M representes the quantity of original image pixels across point; N representes the quantity of the vertical pixel of original image, and Amn representes that coordinate is (m, the value of pixel n) in the original image.

Step S40: the matrix data through two-dimensional dct transform is carried out filtering according to threshold value.

DCT is a kind of real number field conversion, and its transformation kernel is the real number cosine function.After piece image carried out discrete cosine transform, the important visual informations of many relevant images all concentrated in the sub-fraction coefficient of dct transform.Meaning based on dct transform can know that the energy of whole two field picture is assembled to the upper left corner, representes that the closer to the upper left corner frequency is low more.Therefore can the wave filter that will design be calculated based on the parameter of radar system.Remember that fast time-sampling frequency is f _H, slow time-sampling frequency is f _V, the clutter map of intercepting is of a size of the capable N row of M; Here get motion of objects frequency 0.2Hz～5Hz to be detected and be designated as f _Low～f _Hi, real pulse shape f _Sl～f _Sh, then:

Tl＝f _low*M/(f _V/2)

Th＝f _hi*M/(f _V/2)

Tsl＝f _sl*N/(f _H/2)

Tsh-f _sh*N/(f _H/2)

$\left\{\begin{array}{ccc}B(i,j)=0& i<\mathrm{Tlori}>\mathrm{Th},& j<\mathrm{Tslorj}>\mathrm{Tsh}\\ B(i,j)=B(i,j)& \mathrm{Tl}<i<\mathrm{Th},& \mathrm{Tsl}<j<\mathrm{Tsh}\end{array}\right.$

In the formula, Tl is vertical low frequency threshold value, and Th is vertical high frequency threshold value, and Tsl is the horizontal low frequency threshold value, and Tsh is horizontal high frequency threshold value, and (i, j) coordinate is (i, the value of pixel j) to B in the transform domain image behind the expression two-dimensional dct transform.According to following formula the new figure that the DCT coefficient of clutter map constitutes is carried out filtering, specify the DCT coefficient in the area will be, and kept the interior data of frequency range of our needs by zero clearing.Certainly complicated filter only needs calculate the corresponding area size and the position of frequency range of wanting filtering according to following formula if carry out more, then to corresponding D CT coefficient carry out zero clearing perhaps specific modification can form new algorithm, this method applicability is very wide.

Step S50: to carrying out idct transform through filtered matrix data, the reduction clutter map.

In embodiment provided by the invention, to carrying out idct transform through filtered matrix data, the reduction clutter map is realized according to the configurations formula:

$A_{\mathrm{mn}}=\underset{m=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{a}_p{a}_{q}F(p,q)\cos \frac{\mathrm{\&pi;}(2m+1)p}{2M}\cos \frac{\mathrm{\&pi;}(2n+1)q}{2N}$

Wherein, in the formula, p representes the ordinate of transform domain pixel, and q representes the horizontal ordinate of transform domain pixel

$a_p=\left\{\begin{array}{cc}\frac{1}{\sqrt{M}},& p=0\\ \sqrt{\frac{2}{M}},& 1\&le;p\&le;M-1\end{array}\right.,$ $a_q=\left\{\begin{array}{cc}\frac{1}{\sqrt{N}},& q=0\\ \sqrt{\frac{2}{N}},& 1\&le;q\&le;N-1\end{array}\right.,$

0≤p≤M-1,0≤q≤N-1; a _p, a _qIt is one group of coefficient of dct transform.M representes the horizontal ordinate of original image pixel, and n representes the ordinate of original image pixel, and M representes the quantity of original image pixels across point, and N representes the quantity of the vertical pixel of original image, and Amn representes that coordinate is (m, the value of pixel n) in the original image.

Step S50: the clutter map after the reduction is detected.

See also Fig. 3, the schematic flow sheet to the detection of the clutter map after the reduction for the embodiment of the invention provides comprises the steps:

Step S51: the target function value that calculates each point that lists on the clutter map after the reduction.

For the clutter map through DCT filtering and inverse transformation, its size still is N * M, the capable M row of N.Each N that lists point carried out the value of statistical computation objective function, is that example describes here with the covariance: establishing the pixel that i lists is x _i(t) t ∈ [1, N], the then variance on this distance images

$D={\mathrm{\&Sigma;}}_{t=1}^N{[x_i\left(t\right)-\mathrm{Mean}\left(x\right)]}^2,$ In the formula, x _i(t) value of the file pixel on the back image of i.e. reduction on the expression respective distances picture, mean (x) representes the average of these all samples of file pixel.

Be appreciated that except that covariance, can also each kurtosis that is listed as calculate as objective function.What kurtosis was represented is the intensity of one group of DATA DISTRIBUTION, when data sample is very big, when perhaps unknown, should satisfy normal distribution, and its kurtosis is 3; If change because of some reason causes the distribution of data, then kurtosis must depart from 3, so we also can judge whether target through calculating its kurtosis.The calculating formula of kurtosis can adopt the iterative algorithm of simplification:

$K_4^i\left(t\right)=K_4^i(t-1)-\mathrm{\&mu;T}(K_4^i(t-1)-x_i{(t-1)}^4+3)$

Wherein: T is the sampling period; μ gets 60*T, x usually _i(t) value of the file pixel on the back image of i.e. reduction on the expression respective distances picture, The expression kurtosis.This be one iterative, can try to achieve numerical value through this formula traversal sample.。

Step S52: record surpasses the corresponding pixel position of target function value of amplitude threshold.

Choose peaked 80% left and right sides threshold value for adopting variance as the situation of objective function; Promptly calculate the variance yields on all M distance images, get it peaked 80% for threshold value, variance greater than threshold value apart from picture point i; I ∈ [1, M] is recorded in the new array Index.

For the situation that adopts kurtosis as objective function, kurtosis value and 3 is done subtraction, get its absolute value then, to select on all distance images greater than the distance images of maximal value 80%, expression has the target appearance, and abandons less than 80% data.

Step S53: add up above-mentioned pixel position, and calculate its probability density.

In embodiment provided by the invention,, repeatedly repeat the method for above-mentioned pixel position for guaranteeing robustness; Multiplicity is advisable between 4～10 times.The number of times that each i occurs in the statistics Index calculates its probability (occurrence number is divided by multiplicity).

Step S54: select the big point of all density, be the target location.

Select the big N of a density point, be the target location: the probability density among the Index is got peak value, has what peaks what moving targets are just arranged, and its corresponding distance images i, is the distance and position of target.

The size that is appreciated that density depends on artificial selection, in the present embodiment, is preferably more than 80% of maximum density values and is big density.

Below further specify of the present invention through specific embodiment.

For describing conveniently, suppose that it is 512 points that the fast time is gone up every frame storage depth, the distance of correspondence between points is about 8 millimeters; Slow time shaft adopts the SF of 20Hz.We choose 200 frames is a decision window, then whenever can obtain one 200 row at a distance from 10 seconds, 512 row clutter map, be designated as Frame_org, wherein each pixel be designated as f (i, n), wherein i representes the position on the fast time shaft, the span of i is [1,512]; N representes the position on the slow time shaft, and the span of n is [1,200].

See also Fig. 4, go the average processing to obtain new clutter map to each frame, be designated as Frame for what present embodiment provided.This figure is the original image that radar is obtained, and transverse axis is represented fast time shaft, has 512 points here on the transverse axis, and we adopt be a window at 256 on the ordinate, and shared 4 windows splice, altogether 1024 points.

See also Fig. 5, clutter map is carried out the clutter map B that two-dimensional dct transform obtains for provided by the invention.Discrete cosine transform (DCT) is a kind of real number field conversion, and its transformation kernel is the real number cosine function; After piece image carried out discrete cosine transform, the important informations of many relevant images all concentrated in the sub-fraction coefficient of dct transform.The two-dimensional dct of a matrix defines as follows:

$F(p,q)=a_p{a}_q{\mathrm{\&Sigma;}}_{m=0}^{M-1}{\mathrm{\&Sigma;}}_{n=0}^{N-1}{A}_{\mathrm{mn}}\cos \frac{\mathrm{\&pi;}(2m+1)p}{2M}\cos \frac{\mathrm{\&pi;}(2n+1)q}{2N}---\left(1\right)$

Wherein, in the formula, the ordinate of p remarked pixel, the horizontal ordinate of q remarked pixel

$a_p=\left\{\begin{array}{cc}\frac{1}{\sqrt{M}},& p=0\\ \sqrt{\frac{2}{M}},& 1\&le;p\&le;M-1\end{array}\right.,$ $a_q=\left\{\begin{array}{cc}\frac{1}{\sqrt{N}},& q=0\\ \sqrt{\frac{2}{N}},& 1\&le;q\&le;N-1\end{array}\right.,$

0≤p≤M-1,0≤q≤N-1, a _p, a _qBe one group of coefficient of dct transform, m representes the horizontal ordinate of original image pixel, and n representes the ordinate of original image pixel; M representes the quantity of original image pixels across point; N representes the quantity of the vertical pixel of original image, and Amn representes that coordinate is (m, the value of pixel n) in the original image.。

Among the clutter map B, horizontal ordinate and ordinate are the coefficient coordinate of original image behind dct transform, are expressed as the frequency of cosine function.B finds out from clutter map, adopts above-mentioned constructive formula (1) that Frame is carried out two-dimensional dct, obtain having onesize DCT coefficient with former clutter map, and the important information of image all concentrates on the upper left quarter of image.

After obtaining above-mentioned pattern, the matrix data through two-dimensional dct transform is carried out filtering according to threshold value.Specific as follows: as to remember that fast time-sampling frequency is f _H, slow time-sampling frequency is f _V, the clutter map of intercepting is of a size of the capable N row of M; Here get human motion frequency 0.2Hz～5Hz and be designated as f _Low～f _Hi, real pulse shape f _Sl～f _Sh, then:

Tl＝f _low*M/(f _V/2)

Th＝f _hi*M/(f _V/2)

Tsl＝f _si*N/(f _H/2)

Tsh＝f _sh*N/(f _H/2)

$\left\{\begin{array}{ccc}B(i,j)=0& i<\mathrm{Tlori}>\mathrm{Th},& j<\mathrm{Tslorj}>\mathrm{Tsh}\\ B(i,j)=B(i,j)& \mathrm{Tl}<i<\mathrm{Th},& \mathrm{Tsl}<j<\mathrm{Tsh}\end{array}\right.$

In the formula, T1 is vertical low frequency threshold value, and Th is vertical high frequency threshold value, and Tsl is the horizontal low frequency threshold value, and Tsh is horizontal high frequency threshold value.According to following formula the new figure B that the DCT coefficient of clutter map constitutes is carried out filtering, specify the DCT coefficient in the area will be, and kept the interior data of frequency range of our needs by zero clearing.

See also Fig. 6, for the embodiment of the invention provide to carry out the clutter map C that IDCT (DCT inverse transformation) conversion obtains through filtered matrix data.

In the present embodiment, to carrying out idct transform through filtered matrix data, the reduction clutter map is realized according to the configurations formula:

$A_{\mathrm{mn}}=\underset{m=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{a}_p{a}_{q}F(p,q)\cos \frac{\mathrm{\&pi;}(2m+1)p}{2M}\cos \frac{\mathrm{\&pi;}(2n+1)q}{2N}$

Wherein, in the formula, p representes the ordinate of transform domain pixel, and q representes the horizontal ordinate of transform domain pixel

$a_p=\left\{\begin{array}{cc}\frac{1}{\sqrt{M}},& p=0\\ \sqrt{\frac{2}{M}},& 1\&le;p\&le;M-1\end{array}\right.,$ $a_q=\left\{\begin{array}{cc}\frac{1}{\sqrt{N}},& q=0\\ \sqrt{\frac{2}{N}},& 1\&le;q\&le;N-1\end{array}\right.,$

0≤p≤M-1,0≤q≤N-1; a _pA _qIt is one group of coefficient of dct transform.M representes the horizontal ordinate of original image pixel, and n representes the ordinate of original image pixel, and M representes the quantity of original image pixels across point, and N representes the quantity of the vertical pixel of original image, and Amn representes that coordinate is (m, the value of pixel n) in the original image.

Among the clutter map C, the time-domain diagram after the expression filtering reduction can be found out the vestige that motion is arranged in several distance images from clutter map C.According to the clutter map C that obtains; Each column data is added up, and selecting here with the energy is that objective function is judged, calculates its covariance (perhaps ac energy); Can be described as greater than the peak value of maximal value 80% obviously bigger, with the point location target location of satisfying this condition.Except that covariance, can also each kurtosis that is listed as calculate as objective function.What kurtosis was represented is the intensity of one group of DATA DISTRIBUTION, when data sample is very big, when perhaps unknown, should satisfy normal distribution, and its kurtosis is 3; If change because of some reason causes the distribution of data, then kurtosis must depart from 3, so we also can judge whether target through calculating its kurtosis.The calculating formula of kurtosis can adopt the iterative algorithm of simplification:

$K_4^i\left(t\right)=K_4^i(t-1)-\mathrm{\&mu;T}(K_4^i(t-1)-x_i{(t-1)}^4+3)$

Wherein: T is the sampling period, and μ gets 60*T, x usually _i(t) value of the file pixel on the back image of i.e. reduction on the expression respective distances picture,

The expression kurtosis.

See also Fig. 7, the pattern that 4 windows are carried out Conjoint Analysis that provides for the embodiment of the invention.Fig. 7 represent be that each distance images is asked variance after the result, be specially and earlier each window calculated, the criterion that is stitched together the value of 4 windows to the end then.For the sake of assurance, can repeatedly calculate, then the result carried out the Monte Carlo and analyze, choose the several of probability of occurrence maximum, as net result to a plurality of such windows.Show that from Fig. 7 the method that the embodiment of the invention provides can effectively be extracted motion and range information thereof.

See also Fig. 8, the final decision diagram that provides for the embodiment of the invention.Among Fig. 8, horizontal ordinate is represented distance images, and ordinate representes on the corresponding distance images, greater than the number of the point of maximum variance value 80%.From Fig. 5, clearly finding out has 3 moving targets, and its position is respectively 0.3 meter, 1 meter, 3.4 meters, coincide with the experiment situation.

The above only is preferred embodiment of the present invention, is not the present invention is done any pro forma restriction; Though the present invention discloses as above with preferred embodiment; Yet be not in order to limiting the present invention, anyly be familiar with the professional and technical personnel, in not breaking away from technical scheme scope of the present invention; When the technology contents of above-mentioned announcement capable of using is made a little change or is modified to the equivalent embodiment of equivalent variations; In every case be not break away from technical scheme content of the present invention, to any simple modification, equivalent variations and modification that above embodiment did, all still belong in the scope of technical scheme of the present invention according to technical spirit of the present invention.

Claims (9)

1. a pulsed radar moving target detection method is characterized in that, comprises the steps:

The radar system parameter is set to obtain the clutter map size;

According to said radar system calculation of parameter threshold value;

Said clutter map is carried out two-dimensional dct transform;

According to said threshold value the matrix data through two-dimensional dct transform is carried out filtering;

To carrying out idct transform through filtered matrix data, the reduction clutter map; And

Clutter map after the said reduction is detected.

2. pulsed radar moving target detection method according to claim 1 is characterized in that, wherein, the radar system parameter is set to obtain clutter map, comprises the steps:

According to radar performance and sensing range selected frame size;

According to motion of objects characteristic Design sampling system to be detected; And

According to the sampling precision of said sampling system needs, obtain the clutter map size.

3. pulsed radar moving target detection method according to claim 1 and 2 is characterized in that, said threshold value comprises: vertical low frequency threshold value Tl, vertical high frequency threshold value Th, horizontal low frequency threshold value Tsl and horizontal high frequency threshold value Tsh;

Wherein, the computing method of said vertical low frequency threshold value Tl are:

Tl＝f _low*M/(f _V/2)

In the formula, f _LowBe object motion frequency to be detected, M is the line number of said clutter map, f _VBe slow time frequency acquisition;

Wherein, the computing method of said vertical high frequency threshold value Th are:

Th＝f _hi*M/(f _V/2)

In the formula, f _HiBe object motion frequency to be detected, M is the line number of said clutter map, f _VBe slow time frequency acquisition;

Wherein, the computing method of said horizontal low frequency threshold value Tsl are:

Tsl＝f _sl*N/(f _H/2)

In the formula, f _SlBe real pulse shape, N is the columns of said clutter map, f _HBe fast time frequency acquisition; Wherein, the computing method of said horizontal high frequency threshold value Tsh are:

Tsh＝f _sh*N/(f _H/2)

In the formula, f _ShBe real pulse shape, N is the columns of said clutter map, f _HBe fast time frequency acquisition.

4. pulsed radar moving target detection method according to claim 1 is characterized in that, wherein, said clutter map is carried out two-dimensional dct transform realize according to the configurations formula:

$F(p,q)=a_p{a}_q\underset{m=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{A}_{\mathrm{mn}}\cos \frac{\mathrm{\&pi;}(2m+1)p}{2M}\cos \frac{\mathrm{\&pi;}(2n+1)q}{2N}$

Wherein, in the formula, p representes the ordinate of transform domain pixel, and q representes the horizontal ordinate of transform domain pixel

$a_p=\left\{\begin{array}{cc}\frac{1}{\sqrt{M}},& p=0\\ \sqrt{\frac{2}{M}},& 1\&le;p\&le;M-1\end{array}\right.,$ $a_q=\left\{\begin{array}{cc}\frac{1}{\sqrt{N}},& q=0\\ \sqrt{\frac{2}{N}},& 1\&le;q\&le;N-1\end{array}\right.,$

0≤p≤M-1，0≤q≤N-1，

a _pA _qBe one group of coefficient of dct transform, m representes the horizontal ordinate of original image pixel, and n representes the ordinate of original image pixel; M representes the quantity of original image pixels across point; N representes the quantity of the vertical pixel of original image, and Amn representes that coordinate is (m, the value of pixel n) in the original image.

5. according to claim 1 or 3 described pulsed radar moving target detection methods, it is characterized in that, wherein, according to said threshold value the matrix data through two-dimensional dct transform carried out filtering and realize according to the configurations formula:

$\left\{\begin{array}{ccc}B(i,j)=0& i<\mathrm{Tlori}>\mathrm{Th},& j<\mathrm{Tslorj}>\mathrm{Tsh}\\ B(i,j)=B(i,j)& \mathrm{Tl}<i<\mathrm{Th},& \mathrm{Tsl}<j<\mathrm{Tsh}\end{array}\right.$

In the formula, (i, j) coordinate is (i, the value of pixel j) to B in the transform domain image behind the expression two-dimensional dct transform.

6. pulsed radar moving target detection method according to claim 1 is characterized in that, wherein, to carrying out idct transform through filtered matrix data, the reduction clutter map is realized according to the configurations formula:

$A_{\mathrm{mn}}=\underset{p=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\underset{q=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{a}_p{a}_{q}F(p,q)\cos \frac{\mathrm{\&pi;}(2m+1)p}{2M}\cos \frac{\mathrm{\&pi;}(2n+1)q}{2N}$

Wherein, in the formula, p representes the ordinate of transform domain pixel, and q representes the horizontal ordinate of transform domain pixel

$a_p=\left\{\begin{array}{cc}\frac{1}{\sqrt{M}},& p=0\\ \sqrt{\frac{2}{M}},& 1\&le;p\&le;M-1\end{array}\right.,$ $a_q=\left\{\begin{array}{cc}\frac{1}{\sqrt{N}},& q=0\\ \sqrt{\frac{2}{N}},& 1\&le;q\&le;N-1\end{array}\right.,$

0≤p≤M-1,0≤q≤N-1, a _p, a _qIt is one group of coefficient of dct transform.M representes the horizontal ordinate of original image pixel, and n representes the ordinate of original image pixel, and M representes the quantity of original image pixels across point, and N representes the quantity of the vertical pixel of original image, and Amn representes that coordinate is (m, the value of pixel n) in the original image.

7. pulsed radar moving target detection method according to claim 3 is characterized in that, wherein, detection comprises the steps: to the clutter map after the said reduction

Calculate the target function value of each point that lists on the clutter map after the reduction;

Record surpasses the corresponding pixel position of target function value of amplitude threshold;

Add up above-mentioned pixel position, and calculate its probability density; And

Select the big point of all density, be the target location.

8. pulsed radar moving target detection method according to claim 7 is characterized in that, said target function value is a variance yields, and the computing method of said variance yields are:

$D=\underset{t=1}{\overset{N}{\mathrm{\&Sigma;}}}{[x_i\left(t\right)-\mathrm{mean}\left(x\right)]}^2$

In the formula, x _i(t) value of the file pixel on the back image of i.e. reduction on the expression respective distances picture, mean (x) representes the average of these all samples of file pixel.

9. pulsed radar moving target detection method according to claim 7 is characterized in that, said target function value is the kurtosis value of each row, and the computing method of said kurtosis value are:

$K_4^i\left(t\right)=K_4^i(t-1)-\mathrm{\&mu;T}(K_4^i(t-1)-x_i{(t-1)}^4+3)$

In the formula, T is the sampling period, and μ gets 60*T, x _i(t) value of the file pixel on the back image of i.e. reduction on the expression respective distances picture,

The expression kurtosis.

Images