CN109709536A - A kind of SAR moving target detection method based on convolutional neural networks - Google Patents
A kind of SAR moving target detection method based on convolutional neural networks Download PDFInfo
- Publication number
- CN109709536A CN109709536A CN201910065920.4A CN201910065920A CN109709536A CN 109709536 A CN109709536 A CN 109709536A CN 201910065920 A CN201910065920 A CN 201910065920A CN 109709536 A CN109709536 A CN 109709536A
- Authority
- CN
- China
- Prior art keywords
- layer
- convolutional neural
- neural network
- moving target
- target detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013527 convolutional neural network Methods 0.000 title claims abstract description 36
- 238000001514 detection method Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 claims abstract description 14
- 238000012549 training Methods 0.000 claims description 23
- 238000011176 pooling Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000013528 artificial neural network Methods 0.000 abstract description 3
- 230000007547 defect Effects 0.000 abstract description 2
- 239000011159 matrix material Substances 0.000 description 19
- 238000012545 processing Methods 0.000 description 9
- 239000013598 vector Substances 0.000 description 8
- 239000013589 supplement Substances 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 230000001427 coherent effect Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000014509 gene expression Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002592 echocardiography Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
The present invention discloses a kind of SAR moving target detection method based on convolutional neural networks, applied to SAR moving object detection field, target auxiliary data range gate number to be detected can be used to require the defects of high for of the existing technology, method of the invention passes through building convolutional neural networks, and neural network is trained, the angle of the opposite synthetic aperture of total Doppler frequency, target detected according to neural network, is calculated target velocity, to realize that SAR moving-target detects；Has the advantages of detection is realized in the case where target auxiliary data range gate number to be detected is few.
Description
Technical Field
The invention belongs to the technical field of radars, and particularly relates to a moving target detection technology of a Synthetic Aperture Radar (SAR).
Background
The Moving Target Indication (MTI) technology is a key technology of synthetic aperture radar. The moving target detection technology can detect a moving target in a beam irradiation range and estimate moving target parameters such as a moving speed of the moving target.
The conventional MTI method mainly includes Space time adaptive processing (abbreviated as STAP) (reference 1: L.E.Brennan, L.S.Reed.the term of adaptive radar [ J ]. IEEETransmission on Aerosa and Electronic Systems,1973,9(2): 237-. In recent years, researchers have proposed methods for implementing MTI using linear classifiers and polynomial classifiers in machine learning (reference 4: A.E.Khatib, K.Assaleh, H.Mir.Learning-based space-time adaptive Processing [ C ]// International Conference on communications. IEEE,2013:1-4, reference 5: A.E.Khatib, K.Assaleh, H.Mir.Space-time Processing using pattern classification [ J ] IEEE Transactions on Signal Processing,2015,63(3):766 779). However, these methods have the defects that the requirement on the number of available auxiliary data range gates of the target to be detected is high, or the detection can be realized only under the condition that the signal-clutter power of the SAR moving target is relatively high, and the like, so that the application of the existing MTI method is limited, and the requirements of practical application are difficult to meet.
Disclosure of Invention
In order to solve the technical problems, the invention provides a convolutional neural network-based SAR moving target detection method, which can realize accurate SAR moving target detection under the conditions of insufficient available auxiliary data range gate and low signal-to-noise ratio.
The technical scheme adopted by the invention is as follows: a SAR moving target detection method based on a convolutional neural network comprises the following steps:
s1, constructing a convolutional neural network; the method comprises the following steps: the total of 9 convolutional layers, 2 pooling layers, 2 fully-connected layers and 1 final classification layer is recorded as: the 1 st convolutional layer, the 2 nd convolutional layer, the 3 rd pooling layer, the 4 th convolutional layer, the 5 th convolutional layer, the 6 th pooling layer, the 7 th fully-connected layer, the 8 th fully-connected layer and the 9 th final classification layer; the convolution kernel size of the 1 st convolution layer is 5 multiplied by 1, the number of channels is 64, the convolution kernel size of the 2 nd convolution layer is 5 multiplied by 1, the number of channels is 96, the filter size of the 3 rd pooling layer is 3 multiplied by 3, the step length is 1, the convolution kernel size of the 4 th convolution layer is 5 multiplied by 1, the number of channels is 128, the convolution kernel size of the 5 th convolution layer is 5 multiplied by 1, the number of channels is 128, the filter size of the 6 th pooling layer is 3 multiplied by 3, the step length is 2, and the number of output nodes of the 7 th fully-connected layer is 1000; 192 output nodes of the 8 th fully-connected layer; k output nodes of the 9 th final classification layer;
k is a × B, where a is a possible number of total doppler frequencies to be detected and B is a possible number of angles of the target to be detected with respect to the synthetic aperture
S2, constructing SAR moving target detection training dataCollecting; the SAR moving target detection training data set comprises K multiplied by Q multiplied by H training data matrixes X(a,b,q,h)；
Wherein, a represents the serial number of the total Doppler frequency; b represents the number of the angle of the target relative to the synthetic aperture; q denotes the sequence number of the auxiliary data, Q1, 2.., Q denotes the total number of auxiliary data from the gate; h denotes the sequence number of the training data for target amplitude construction, and H1, 2.
S3, training the convolutional neural network constructed in the step S1 according to the SAR moving target detection training data set constructed in the step S2; specifically, the formula Θ ═ argminJ (Θ) is used to obtain the weight and bias parameters of the convolutional neural network, where Θ is all the parameters of the convolutional neural network, and J (Θ) is the cross entropy loss function.
Wherein p (k | X)(k,q,h)(ii) a Θ) is the output data of the last layer of the convolutional neural network.
And S4, detecting the SAR echo data to be detected according to the convolutional neural network trained in the step S3 to obtain a target speed. Specifically, a and b are respectively calculated according to the following formula;
b＝mod((k+1),B)
and calculating the speed of the corresponding SAR moving target according to the a and the b.
The invention has the beneficial effects that: the SAR moving target detection method based on the convolutional neural network can realize SAR moving target detection under the condition that only 8 auxiliary data range gates are provided, and in an application environment with a low signal-to-noise ratio, moving target echoes are submerged in clutter and noise, so that the accuracy of more than 90% can be still obtained; compared with the existing method, the method has less requirements on the number of the auxiliary range gates and can show robustness in heterogeneous environment; the method of the invention has the following advantages:
1. useful features are extracted layer by utilizing a deep learning neural network, and a pooling layer is used, so that network parameters are reduced, and the overfitting degree of a model is reduced;
2. a large amount of training data is utilized, each class generates the same amount of training data, and the problem of class imbalance is avoided;
3. the auxiliary data is only used as an interference background, so that the requirement of the auxiliary data on the number of the distance gates and the influence of the heterogeneity of the auxiliary data are greatly reduced.
Drawings
FIG. 1 is a flow chart of a protocol provided in the practice of the present invention;
fig. 2 is a diagram of a convolutional neural network architecture provided in the practice of the present invention.
Detailed Description
For the convenience of describing the present invention, the following terms are first defined:
define 1, space-time observation matrix
The space-time observation matrix X is a two-dimensional matrix, wherein the result of SAR echo signal distance compression is stored, each row of the two-dimensional matrix represents each pulse, each column represents each antenna channel, and the mathematical expression of the mth row and the nth column of elements of the two-dimensional matrix is
The first term in the formula (1) represents the echo of the SAR moving target signal, and the second term χm,nRepresenting clutter and noise interference, M1., M, N1., N, M is the number of transmitted pulses in a coherent processing interval, N is the number of antenna channels, M, N are the pulse and antenna indices, respectively, α is the amplitude of the moving target,is a random phase,/n＝Ln/λ，LnIs the distance of the nth antenna from the 1 st antenna, λ is the signal wavelength at which the radar system operates, θ is the angle of the target relative to the synthetic aperture, ftIs the total Doppler frequency, including the Doppler frequency f caused by the motion of the radar platformdAnd Doppler frequency f caused by object motionvCan be expressed as
ft＝fdsin(θ)+fv(2)
Where φ is the angle of the azimuth velocity of the moving target relative to the target velocity, vtIs the speed of the moving object, vpIs the speed of the radar platform, frIs the pulse repetition frequency of the radar system.
Definition 2, convolutional neural network
The convolutional neural network is an end-to-end classifier and comprises a convolutional layer, a pooling layer, a full connection layer, a softmax layer and a classification layer.
Definition 3, convolution layer
The convolutional layer performs both linear convolution and non-linear processing operations, and can be expressed as
Wherein,respectively output data and input data of the layer, gammac(. cndot.) is a non-linear activation operator,respectively corresponding weight parameters and bias parameters,is the channel index of the input and output, Fc×Fc,ξ,η＝1,...,FcIs the filter size, ScIs the step size of the convolutional layer, representing the spacing between two adjacent input sub-regions.
Definition 4, pooling layer
Operations performed by the pooling layer may be represented as
Wherein,respectively output data and input data of the layer, Fp×FpIs the filter size of the input subregion, SpIs the step size of the pooling layer, representing the separation between two adjacent input sub-regions.
Definition 5, fully connected layer
The operation performed by the fully connected layer may be represented as
Wherein,respectively output data and input data of the layer, gammaf(-) is the nonlinear function operator of the fully-connected layer,is the weight vector and offset connecting the mth element of the layer input vector and output vector,<·>representing the inner product of two vectors.
Definition 6, Final Classification level
The final classification layer performs the following calculations
Wherein, yk,yfRespectively output data and input data of the layer, wk,bkIs the weight vector and offset that connects the kth element of the layer input vector and output vector.
In order to facilitate the understanding of the technical contents of the present invention by those skilled in the art, the present invention will be further explained with reference to the accompanying drawings.
Fig. 1 is a flowchart of an SAR moving target detection method based on a convolutional neural network, which is specifically implemented as follows:
step 1: constructing convolutional neural networks
A convolutional neural network is constructed, the network structure is shown in fig. 2, and the network comprises 4 convolutional layers, 2 pooling layers, 2 fully-connected layers and 1 final classification layer. The convolution kernel size of the convolution layer of the 1 st layer is 5 multiplied by 1, the number of channels is 64, all 0 supplements are used, the step length is 1, the input matrix of the layer is each training data, namely the input size is 438 multiplied by 3 multiplied by 2; the convolution kernel size of the convolution layer of the 2 nd layer is 5 multiplied by 1, the number of channels is 96, all 0 supplements are used, the step length is 1, the input matrix of the layer is the output matrix of the previous layer, namely the input size is 438 multiplied by 3 multiplied by 64; the filter size of the 3 rd pooling layer is 3 × 3, the step length is 1, the input matrix of the layer is the output matrix of the previous layer, namely the input size is 438 × 3 × 96; the convolution kernel size of the convolution layer of the 4 th layer is 5 multiplied by 1, the number of channels is 128, all 0 supplements are used, the step length is 1, the input matrix of the layer is the output matrix of the previous layer, namely the input size is 438 multiplied by 3 multiplied by 96; the convolution kernel size of the convolution layer of the 5 th layer is 5 multiplied by 1, the number of channels is 128, all 0 supplements are used, the step length is 1, the input matrix of the layer is the output matrix of the previous layer, namely the input size is 438 multiplied by 3 multiplied by 128; the filter size of the 6 th pooling layer is 3 × 3, the step length is 2, the input matrix of the layer is the output matrix of the previous layer, namely the input size is 438 × 3 × 128; the number of input nodes of the 7 th layer full connection layer is 219 multiplied by 2 multiplied by 128, and the number of output nodes is 1000; the number of input nodes of the 8 th full-connection layer is 1000, and the number of output nodes is 192; the number of input nodes of the 9 th final classification layer is 192, and the number of output nodes is 15.
Step 2: initializing radar system parameters
Initializing radar system parameters: initializing radar system parameters: signal wavelength at which the radar system operates: lambda; number of transmitted pulses within one coherent processing interval: m; number of antenna channels: n; radar system pulse repetition frequency: f. ofr(ii) a Speed of radar platform: v. ofp. In this embodiment, the working signal wavelength λ of the radar system is 0.0312m, and other parameters are detailed in attached table 1.
TABLE 1 list of system parameters of measured data
| Parameter(s) | Measured data |
| Antenna with a shieldNumber of channels N | 3 |
| Number of emitted pulses M within a coherent processing interval | 438 |
| Pulse repetition frequency fr(Hz) | 2.1716e+03 |
| Radar platform velocity vp(m/s) | 104.2616 |
And step 3: constructing SAR moving target detection training data set
A SAR moving target detection training data set is enhanced by using possible combinations of 15 total Doppler frequencies and angles of targets relative to a synthetic aperture, limited 8 auxiliary data and 201 target amplitudes, wherein one training data setIs a two-dimensional data matrix of 438 rows and 3 columns, the two-dimensional matrixThe mth row and nth column elements of (1):
wherein, superscripts a, B, q, h respectively represent different total doppler frequencies, angles of the target relative to the synthetic aperture, auxiliary data range gates and amplitudes of the artificially constructed target, and K ═ a × B, K represents combinations of different total doppler frequencies and angles of the target relative to the synthetic aperture, and correspondence of K, a, B is given below
b＝mod((k+1),B) (11)
Wherein,this indicates rounding down, mod (·) indicates the remainder of the calculation, and when B ≠ 0, B ═ mod [ (k +1), B]When B is 0, B is B.
In this embodiment, a is 5, B is 3, a is 1,2,3,4,5, B is 1,2,3, k is 0,1,2, 14, q is 1,2, 8, h is 1, 2.
In the formula (9), random phaseObeying a uniform distribution over [0, 2 π) and varying with a, b, q, h, α(a,b,h)Is the h-th target amplitude of the k-th class (k is obtained from a, b according to the equations (10) and (11)), andwhereinRespectively, the upper and lower bounds of the amplitude of the class k training data.
And 4, step 4: training convolutional neural networks
Training the convolutional neural network by using the training data set in step 3, and obtaining the weight and bias parameters of the convolutional neural network by using a formula Θ ═ argminJ (Θ), wherein Θ is all parameters including the convolutional neural network, and J (Θ) is a cross entropy loss function, that is, J (Θ) is a cross entropy loss functionIn the formula, p (k | X)(k,q,h)；Θ) is the output of the last layer of the convolutional neural network.
By solving the optimization problem by using a gradient descent and back propagation method, the method can obtainWhere i 1, 2., 400 denotes the number of iterations of gradient descent, the initial value Θ (0) of the parameter vector is set randomly, η (i) is the learning rate of the ith iteration,is the gradient calculation.
And 5: detecting SAR moving target
Taking an SAR echo data matrix to be detected as the input of a network, and then carrying out test processing to obtain a kth class of a corresponding class number, whereinWhen K is 0,1, 2., K-1, the corresponding angle between the a-th total doppler frequency and the b-th target relative to the synthetic aperture can be obtained according to the expressions (10) and (11), and finally the corresponding speed of the SAR moving target can be obtained according to the expressions (2), (3) and (4).
It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (8)
1. A SAR moving target detection method based on a convolutional neural network is characterized by comprising the following steps:
s1, constructing a convolutional neural network; the method comprises the following steps: the total of 9 convolutional layers, 2 pooling layers, 2 fully-connected layers and 1 final classification layer is recorded as: the 1 st convolutional layer, the 2 nd convolutional layer, the 3 rd pooling layer, the 4 th convolutional layer, the 5 th convolutional layer, the 6 th pooling layer, the 7 th fully-connected layer, the 8 th fully-connected layer and the 9 th final classification layer; the convolution kernel size of the 1 st convolution layer is 5 multiplied by 1, the number of channels is 64, the convolution kernel size of the 2 nd convolution layer is 5 multiplied by 1, the number of channels is 96, the filter size of the 3 rd pooling layer is 3 multiplied by 3, the step length is 1, the convolution kernel size of the 4 th convolution layer is 5 multiplied by 1, the number of channels is 128, the convolution kernel size of the 5 th convolution layer is 5 multiplied by 1, the number of channels is 128, the filter size of the 6 th pooling layer is 3 multiplied by 3, the step length is 2, and the number of output nodes of the 7 th fully-connected layer is 1000; 192 output nodes of the 8 th fully-connected layer; k output nodes of the 9 th final classification layer;
s2, constructing an SAR moving target detection training data set;
s3, training the convolutional neural network constructed in the step S1 according to the SAR moving target detection training data set constructed in the step S2;
and S4, detecting the SAR echo data to be detected according to the convolutional neural network trained in the step S3 to obtain a target speed.
2. The method as claimed in claim 1, wherein K is axb, a is the possible number of total doppler frequencies to be detected, and B is the possible number of angles of the target to be detected with respect to the synthetic aperture in step S1.
3. The SAR moving target detection method based on convolutional neural network as claimed in claim 2, wherein the SAR moving target detection training data set of step S2 includes K × Q × H training data matrices X(a,b,q,h)；
Wherein, a represents the serial number of the total Doppler frequency; b represents the number of the angle of the target relative to the synthetic aperture; q denotes the sequence number of the auxiliary data, Q1, 2.., Q denotes the total number of auxiliary data from the gate; h denotes the sequence number of the training data for target amplitude construction, and H1, 2.
4. The method according to claim 3, wherein the step S3 specifically adopts a formula Θ ═ arg min J (Θ) to obtain the weight and bias parameters of the convolutional neural network, where Θ is all the parameters including the convolutional neural network, and J (Θ) is a cross entropy loss function.
5. The SAR moving target detection method based on convolutional neural network as claimed in claim 4,wherein k represents the combination serial number of different total Doppler frequencies and the angle of the target relative to the synthetic aperture; k is 0,1,., K-1, Q denotes the serial number of the auxiliary data, Q is 1,2,., Q; h represents the sequence number of the training data constructed by the target amplitude, and H is 1, 2. p (k | X)(k,q,h)(ii) a Θ) is the output data of the last layer of the convolutional neural network.
6. The SAR moving target detection method based on convolutional neural network as claimed in claim 5, wherein step S4 further comprises: and obtaining the target speed according to the calculated total Doppler frequency corresponding to the SAR to be detected and the angle of the target relative to the synthetic aperture.
7. The SAR moving target detection method based on the convolutional neural network as claimed in claim 6, wherein the calculation formula of the sequence number a of the total Doppler frequency is as follows:
wherein,indicating a rounding down.
8. The SAR moving target detection method based on the convolutional neural network as claimed in claim 6, wherein the sequence number b of the angle of the target relative to the synthetic aperture is calculated by the following formula:
b＝mod((k+1),B)
where mod (-) represents the remainder operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910065920.4A CN109709536A (en) | 2019-01-24 | 2019-01-24 | A kind of SAR moving target detection method based on convolutional neural networks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910065920.4A CN109709536A (en) | 2019-01-24 | 2019-01-24 | A kind of SAR moving target detection method based on convolutional neural networks |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109709536A true CN109709536A (en) | 2019-05-03 |
Family
ID=66261740
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910065920.4A Pending CN109709536A (en) | 2019-01-24 | 2019-01-24 | A kind of SAR moving target detection method based on convolutional neural networks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109709536A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111123257A (en) * | 2019-12-30 | 2020-05-08 | 西安电子科技大学 | Radar moving target multi-frame joint detection method based on graph space-time network |
| CN112183534A (en) * | 2020-10-07 | 2021-01-05 | 西安电子科技大学 | Moving target intelligent combined detection method based on video synthetic aperture radar |
| CN112180338A (en) * | 2020-06-10 | 2021-01-05 | 四川九洲电器集团有限责任公司 | Holographic digital array radar target quantity estimation method and system |
| CN112748432A (en) * | 2020-12-25 | 2021-05-04 | 中国科学院空天信息创新研究院 | Method and device for alternately executing strip mode and wide area MTI mode by airborne SAR |
| CN113240047A (en) * | 2021-06-02 | 2021-08-10 | 西安电子科技大学 | SAR target recognition method based on component analysis multi-scale convolutional neural network |
| CN113253272A (en) * | 2021-07-15 | 2021-08-13 | 中国人民解放军国防科技大学 | Target detection method and device based on SAR distance compressed domain image |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6870501B2 (en) * | 2001-06-26 | 2005-03-22 | Raytheon Company | Digital radio frequency tag |
| DE502004001548D1 (en) * | 2003-06-23 | 2006-11-02 | Eads Deutschland Gmbh | SIGNAL EVALUATION METHOD IN A SAR / MTI PULSE RADAR SYSTEM |
| CN106228124A (en) * | 2016-07-17 | 2016-12-14 | 西安电子科技大学 | SAR image object detection method based on convolutional neural networks |
| CN106443596A (en) * | 2016-09-12 | 2017-02-22 | 电子科技大学 | SVM (support vector machine) space-time adaptive processing method |
| CN106597425A (en) * | 2016-11-18 | 2017-04-26 | 中国空间技术研究院 | Radar object positioning method based on machine learning |
| CN107015214A (en) * | 2017-06-06 | 2017-08-04 | 电子科技大学 | A kind of space-time adaptive processing method based on management loading |
| CN107256396A (en) * | 2017-06-12 | 2017-10-17 | 电子科技大学 | Ship target ISAR characteristics of image learning methods based on convolutional neural networks |
| CN107784320A (en) * | 2017-09-27 | 2018-03-09 | 电子科技大学 | Radar range profile's target identification method based on convolution SVMs |
| CN108280412A (en) * | 2018-01-12 | 2018-07-13 | 西安电子科技大学 | High Resolution SAR image object detection method based on structure changes CNN |
| CN108872988A (en) * | 2018-07-12 | 2018-11-23 | 南京航空航天大学 | A kind of inverse synthetic aperture radar imaging method based on convolutional neural networks |
| CN108960190A (en) * | 2018-07-23 | 2018-12-07 | 西安电子科技大学 | SAR video object detection method based on FCN Image Sequence Model |
| CN109086799A (en) * | 2018-07-04 | 2018-12-25 | 江苏大学 | A kind of crop leaf disease recognition method based on improvement convolutional neural networks model AlexNet |
-
2019
- 2019-01-24 CN CN201910065920.4A patent/CN109709536A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6870501B2 (en) * | 2001-06-26 | 2005-03-22 | Raytheon Company | Digital radio frequency tag |
| DE502004001548D1 (en) * | 2003-06-23 | 2006-11-02 | Eads Deutschland Gmbh | SIGNAL EVALUATION METHOD IN A SAR / MTI PULSE RADAR SYSTEM |
| CN106228124A (en) * | 2016-07-17 | 2016-12-14 | 西安电子科技大学 | SAR image object detection method based on convolutional neural networks |
| CN106443596A (en) * | 2016-09-12 | 2017-02-22 | 电子科技大学 | SVM (support vector machine) space-time adaptive processing method |
| CN106597425A (en) * | 2016-11-18 | 2017-04-26 | 中国空间技术研究院 | Radar object positioning method based on machine learning |
| CN107015214A (en) * | 2017-06-06 | 2017-08-04 | 电子科技大学 | A kind of space-time adaptive processing method based on management loading |
| CN107256396A (en) * | 2017-06-12 | 2017-10-17 | 电子科技大学 | Ship target ISAR characteristics of image learning methods based on convolutional neural networks |
| CN107784320A (en) * | 2017-09-27 | 2018-03-09 | 电子科技大学 | Radar range profile's target identification method based on convolution SVMs |
| CN108280412A (en) * | 2018-01-12 | 2018-07-13 | 西安电子科技大学 | High Resolution SAR image object detection method based on structure changes CNN |
| CN109086799A (en) * | 2018-07-04 | 2018-12-25 | 江苏大学 | A kind of crop leaf disease recognition method based on improvement convolutional neural networks model AlexNet |
| CN108872988A (en) * | 2018-07-12 | 2018-11-23 | 南京航空航天大学 | A kind of inverse synthetic aperture radar imaging method based on convolutional neural networks |
| CN108960190A (en) * | 2018-07-23 | 2018-12-07 | 西安电子科技大学 | SAR video object detection method based on FCN Image Sequence Model |
Non-Patent Citations (6)
| Title |
|---|
| SHAO, C. QU AND J. LI: "A performance analysis of convolutional neural network models in SAR target recognition", 《2017 SAR IN BIG DATA ERA: MODELS, METHODS AND APPLICATIONS (BIGSARDATA), BEIJING》 * |
| Z. LIN, K. JI, M. KANG, X. LENG AND H. ZOU: "Deep Convolutional Highway Unit Network for SAR Target Classification With Limited Labeled Training Data", 《IEEE GEOSCIENCE AND REMOTE SENSING LETTERS》 * |
| Z. LIU, D. K. C. HO, X. XU AND J. YANG: "Moving Target Indication Using Deep Convolutional Neural Network", 《IEEE ACCESS》 * |
| 李健伟,曲长文,彭书娟,邓兵: "基于卷积神经网络的SAR图像舰船目标检测", 《系统工程与电子技术》 * |
| 肖定坤: "基于深度网络的SAR图像目标检测技术研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 许强,李伟,PIERRE LOUMBI: "深度卷积神经网络在SAR自动目标识别领域的应用综述", 《电讯技术》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111123257A (en) * | 2019-12-30 | 2020-05-08 | 西安电子科技大学 | Radar moving target multi-frame joint detection method based on graph space-time network |
| CN111123257B (en) * | 2019-12-30 | 2023-03-28 | 西安电子科技大学 | Radar moving target multi-frame joint detection method based on graph space-time network |
| CN112180338A (en) * | 2020-06-10 | 2021-01-05 | 四川九洲电器集团有限责任公司 | Holographic digital array radar target quantity estimation method and system |
| CN112180338B (en) * | 2020-06-10 | 2022-03-01 | 四川九洲电器集团有限责任公司 | Holographic digital array radar target quantity estimation method and system |
| CN112183534A (en) * | 2020-10-07 | 2021-01-05 | 西安电子科技大学 | Moving target intelligent combined detection method based on video synthetic aperture radar |
| CN112183534B (en) * | 2020-10-07 | 2023-05-23 | 西安电子科技大学 | Moving target intelligent joint detection method based on video synthetic aperture radar |
| CN112748432A (en) * | 2020-12-25 | 2021-05-04 | 中国科学院空天信息创新研究院 | Method and device for alternately executing strip mode and wide area MTI mode by airborne SAR |
| CN112748432B (en) * | 2020-12-25 | 2023-09-29 | 中国科学院空天信息创新研究院 | Method and device for alternately executing stripe mode and wide area MTI mode by airborne SAR |
| CN113240047A (en) * | 2021-06-02 | 2021-08-10 | 西安电子科技大学 | SAR target recognition method based on component analysis multi-scale convolutional neural network |
| CN113240047B (en) * | 2021-06-02 | 2022-12-02 | 西安电子科技大学 | SAR target recognition method based on component analysis multi-scale convolutional neural network |
| CN113253272A (en) * | 2021-07-15 | 2021-08-13 | 中国人民解放军国防科技大学 | Target detection method and device based on SAR distance compressed domain image |
| CN113253272B (en) * | 2021-07-15 | 2021-10-29 | 中国人民解放军国防科技大学 | Target detection method and device based on SAR distance compressed domain image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109709536A (en) | A kind of SAR moving target detection method based on convolutional neural networks | |
| JP6699904B2 (en) | Radar device and radar signal processing method thereof | |
| CN109100718A (en) | Sparse aperture ISAR self-focusing and transverse calibration method based on Bayesian learning | |
| CN105699952B (en) | Double quantile methods of estimation of sea clutter K profile shape parameters | |
| CN108919193A (en) | A kind of parameter agile radar signal sorting method excavated based on sequence fragment | |
| CN104316914B (en) | Radar target self-adaptation detection method depending on shape parameters | |
| CN103713285A (en) | Distributed meter wave array radar angle measuring method based on information fusion | |
| CN106483516A (en) | Radar clutter space-time adaptive processing method based on priori | |
| CN112612006B (en) | Deep learning-based non-uniform clutter suppression method for airborne radar | |
| CN110146873B (en) | Target position and speed estimation method of distributed non-coherent radar | |
| CN106970370A (en) | Radar Multi Target tracking optimization method based on chaotic neural network | |
| CN110308432A (en) | A kind of radar self-adaption waveform selection Activity recognition method neural network based | |
| CN109919082A (en) | Modal identification method based on LSTM and EMD | |
| CN104535973A (en) | Target detection method by use of airborne early warning radar | |
| JP2018205175A (en) | Radar device and radar signal processing method thereof | |
| Bhatti et al. | Radar signals intrapulse modulation recognition using phase-based stft and bilstm | |
| Wang et al. | Model-and-data-driven method for radar highly maneuvering target detection | |
| CN110764084A (en) | Radar detection method for shielding target under complex terrain | |
| CN106199552A (en) | A kind of packet generalized likelihood test method under local uniform sea clutter background | |
| CN111650574B (en) | Underwater space-time self-adaptive processing method and system based on sparse recovery | |
| CN109541567B (en) | High-speed maneuvering target detection method based on deep learning | |
| CN112213706B (en) | Logic TBD detection method for multi-dimensional digital wave gate interframe progressive association | |
| CN108983160A (en) | Compressed sensing Radar Multi Target method for parameter estimation in the case of a kind of low signal-to-noise ratio | |
| CN114740446A (en) | Method for detecting weak moving target in sea clutter based on stochastic resonance | |
| CN113281776A (en) | Laser radar target intelligent detector for complex underwater dynamic target |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190503 |
|
| RJ01 | Rejection of invention patent application after publication |