CN110109110B - HRRP target identification method based on priori optimal variation self-encoder - Google Patents
HRRP target identification method based on priori optimal variation self-encoder Download PDFInfo
- Publication number
- CN110109110B CN110109110B CN201910346042.3A CN201910346042A CN110109110B CN 110109110 B CN110109110 B CN 110109110B CN 201910346042 A CN201910346042 A CN 201910346042A CN 110109110 B CN110109110 B CN 110109110B
- Authority
- CN
- China
- Prior art keywords
- layer perceptron
- data
- perceptron
- training
- imaging data
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
- G01S13/9021—SAR image post-processing techniques
- G01S13/9027—Pattern recognition for feature extraction
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
- G01S13/9094—Theoretical aspects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/417—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section involving the use of neural networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/418—Theoretical aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Theoretical Computer Science (AREA)
- Electromagnetism (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Radar Systems Or Details Thereof (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a radar high-resolution range profile target recognition method based on a priori optimal variation self-encoder, which mainly solves the problem of low recognition rate in the prior art and comprises the following implementation steps: 1) Acquiring radar high-resolution distance imaging data, and dividing the imaging data into a training sample set and a test sample set; 2) Preprocessing radar high-resolution range profile data, and constructing a priori optimal variation self-encoder consisting of two perceptrons; 3) Training the priori optimal variation self-encoder by using training set data to obtain a trained first multi-layer perceptron and a trained second multi-layer perceptron; 4) Extracting hidden variable features a corresponding to a training set to train a Support Vector Machine (SVM); 5) Extracting hidden variable features b corresponding to the test set, and completing target identification of the test set through a trained Support Vector Machine (SVM). The invention obviously improves the recognition rate and the robustness of recognizing the noisy samples, and can be used for environment detection and track tracking.
Description
Technical Field
The invention belongs to the technical field of radars, and particularly relates to a radar high-resolution range profile HRRP target recognition method which can be used for environment detection and track tracking.
Background
The range resolution of the radar is proportional to the received pulse width after matched filtering, and the range unit length of the radar transmitting signal meets the following conditions:
Δr is the distance unit length of the radar transmit signal, c is the speed of light, τ is the pulse width of the matched receive, and B is the bandwidth of the radar transmit signal. The radar range resolution is relative to the observed target, when the observed target has a dimension L along the radar sight line direction, if L < DeltaR, the corresponding radar echo signal width is approximately the same as the radar emission pulse width, and is commonly called as a 'point' target echo, and the radar is a low-resolution radar; if L > ΔR, then targetThe echoes become a "one-dimensional range profile" extending in distance according to the characteristics of the target, such radars being high resolution radars, where the symbol < means much less than and > means much greater than.
The working frequency of the high-resolution radar is located in a high-frequency area relative to a general target, a broadband coherent signal is transmitted, and the radar receives echo data by transmitting electromagnetic waves to the target. Typically the echo characteristics are calculated using a simplified scatter point model, i.e. using a first approximation of the born ignoring multiple scatter.
The fluctuation and peak appearing in the high-resolution radar echo reflect the distribution condition of radar scattering sectional areas RCS of a target such as a nose, a wing, a tail rudder, an air inlet hole, an engine and the like along the radar sight RLOS when the target scatters at a certain radar view angle, and reflect the relative geometrical relationship of scattering points in the radial direction, which is commonly called as a high-resolution range profile HRRP. Therefore, the HRRP sample contains important structural features of the target, which is valuable for target identification and classification.
At present, many target recognition methods for high-resolution range profile HRRP data have been developed, including direct classification of targets using a more traditional support vector machine, classification using a convolutional neural network based on a time domain, and feature extraction methods for limiting boltzmann machines, but these methods have common disadvantages: the accuracy of target identification is not high.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a radar high-resolution range profile HRRP target recognition method based on a priori optimal variation self-encoder so as to improve the recognition accuracy of targets.
In order to achieve the above purpose, the technical scheme of the invention comprises the following steps:
(1) Extracting imaging data x of a radar echo of a Q-type target along a distance dimension on a radar line, and randomly dividing the Q-type high-resolution distance imaging data into a training sample set and a test sample set; wherein Q is a positive integer greater than 0;
(2) Carrying out mean normalization preprocessing on imaging data to obtain preprocessed data x';
(3) Constructing a priori optimal variation self-coding machine consisting of two perceptron machines;
(3a) Setting a first multi-layer perceptron which comprises two hidden layers, wherein each hidden layer is provided with 400 nodes, the input dimension of the perceptron is 256, and the output dimension is 60;
(3b) Setting a second multi-layer perceptron which comprises two hidden layers, wherein each hidden layer is provided with 400 nodes, the input dimension of the perceptron is 30, and the output dimension is 256;
(3c) Performing re-parametric processing on the output of the first multi-layer perceptron, and taking the output of the first multi-layer perceptron processed by the re-parametric processing as the input of the second multi-layer perceptron to form a priori optimal variation self-encoder;
(4) Training the priori optimal variation self-encoder by using training set data x', and learning parameters of the first multi-layer perceptron and the second multi-layer perceptron to obtain a trained first multi-layer perceptron and a trained second multi-layer perceptron;
(5) Sequentially inputting training set data and test set data into a trained first multi-layer perceptron, and respectively extracting hidden variable features a corresponding to the training set and hidden variable features b corresponding to the test set;
(6) Inputting the hidden variable features a corresponding to the training set into a Support Vector Machine (SVM), training a support vector machine classifier, and classifying the hidden variable features b corresponding to the testing set by using the trained support vector machine classifier to obtain a target recognition result.
Compared with the prior art, the invention has the following advantages:
first, robustness is strong: according to the method, probability modeling is adopted, energy normalization preprocessing is carried out on the data, and noise of the data is considered in a distribution function of the model, so that the robustness of the model can be effectively improved.
Second, the target recognition rate is high: the traditional target recognition method for the high-resolution range profile HRRP data generally only uses a traditional classifier to directly classify the original data to obtain a recognition result, and does not extract high-dimensional features of the data, so that the recognition rate is not high.
Drawings
FIG. 1 is a flow chart of an implementation of the present invention;
Detailed Description
The invention is described in further detail below with reference to the drawings and the detailed description.
Referring to fig. 1, the steps for implementing the present invention are as follows:
and step 1, acquiring radar high-resolution distance imaging data, and dividing the imaging data into a training sample set and a test sample set.
(1a) The high-resolution radar transmits electromagnetic waves to the Q-class targets, receives imaging data x of the distribution condition of the scattering cross section area RCS of the Q-class targets along the radar sight RLOS, and sequentially marks the echo data as class 1 high-resolution distance imaging data, class 2 high-resolution distance imaging data, … and class Q high-resolution distance imaging data;
(1b) And extracting P training sample sets and A test sample sets from the Q-class high-resolution distance imaging data in a random extraction mode of the track segment, wherein each training sample or test sample contains N distance units, and a P multiplied by N dimension training sample set matrix and an A multiplied by N dimension test sample set matrix are obtained.
And 2, preprocessing radar high-resolution range profile data.
The imaging data x is preprocessed, and the existing preprocessing method comprises preprocessing of energy normalization and alignment. In this embodiment, energy normalization processing is used, but not limited to, to obtain normalized data x':
wherein I 2 Representing the data x 'after the mean normalization processing to obtain a binary norm, wherein the data x' is a P multiplied by N dimensional matrix, P represents the total number of training samples contained in the training sample set, and N represents P training samplesThe total number of distance units contained in each type of high-resolution distance imaging data respectively.
And 3, constructing a priori optimal variation self-encoding machine consisting of two perceptron machines.
(3a) Setting a first multi-layer perceptron, wherein the perceptron comprises two hidden layers, each hidden layer is provided with l nodes, the input dimension of the perceptron is 256, and the output dimension is k; in this embodiment, the value of l is 400, and the value of k is 60;
(3b) Setting a second multi-layer perceptron which comprises two hidden layers, wherein each hidden layer is provided with j nodes respectively, and the input dimension of the perceptron is as follows
The output dimension is 256; in this embodiment, the value of j is 400;
(3c) Inputting x' into a first multi-layer perceptron to obtain the output of the first multi-layer perceptron;
(3d) Equally dividing the output of the first multi-layer perceptron into two parts, respectively marking the two parts as mu and sigma;
(3e) Sampling data from a gaussian distribution with a mean value of 0 and a variance of 1, and marking the data as epsilon;
(3f) Calculating an input z of the second multi-layer perceptron from the results of (3 d) and (3 e):
z＝μ+ε×σ。
and 4, training the priori optimal variation self-encoder by using training set data x' to obtain a trained first multi-layer perceptron and a trained second multi-layer perceptron.
(4a) Inputting training set data x' into a first multi-layer perceptron, and calculating the conditional log likelihood lnp corresponding to the x θ Lower variation bound of (x' |y):
wherein phi represents the parameter of the first multi-layer perceptron, theta 1 The condition prior distribution full statistic quantity of the output of the first multi-layer perceptron is represented by theta 2 The parameters of the second multi-layer perceptron, y representing the x' correspondenceTarget class, p θ Representing a conditional distribution probability determined based on the second multi-layer perceptron output,
representing a variable lower bound function symbol;
(4b) Maximizing a variation lower bound formula by using a batch random gradient descent method, and distributing the priori optimal variation from the parameter phi of a first multi-layer perceptron of the encoder, wherein the priori distribution of the condition output by the first multi-layer perceptron is sufficient in statistics theta 1 And a parameter θ of the second multilayer perceptron 2 And carrying out iterative updating for 500 times to obtain the trained first multi-layer perceptron and the trained second multi-layer perceptron.
And 5, extracting hidden variable features a corresponding to the training set and hidden variable features b corresponding to the test set.
And sequentially inputting the training set data and the testing set data into a trained first multi-layer perceptron, and respectively extracting hidden variable features a corresponding to the training set and hidden variable features b corresponding to the testing set.
And 6, completing a target recognition task of the test set by using a Support Vector Machine (SVM).
Inputting the hidden variable features a corresponding to the training set into a Support Vector Machine (SVM), training a support vector machine classifier, and classifying the hidden variable features b corresponding to the testing set by using the trained support vector machine classifier to obtain a target recognition result.
The effect of the invention can be further verified and demonstrated by the following simulation experiment.
Experimental conditions
1. The data used in the experiment are HRRP measured data of high-resolution range profile of 3 types of aircrafts, the 3 types of aircrafts are respectively in a trophy state, an ampere 26 and a jacob 42, the 3 types of high-resolution range imaging data obtained are respectively the high-resolution range imaging data of the trophy-state aircrafts, the high-resolution range imaging data of the ampere 26 aircraft and the high-resolution range imaging data of the jacob 42, the 2 nd and 5 th flight tracks, the 6 th and 7 th flight tracks of the trophy state and the 5 th and 6 th flight tracks of the "ampere-26" are extracted as training samples, and the rest of imaging data are used as test sample sets.
2. Respectively adding corresponding category labels to all high-resolution distance imaging data in the training sample set and the test sample set; the training sample set comprises 140000 training samples, the test sample set comprises 5200 test samples, wherein the training samples comprise 52000 types of 1 high-resolution imaging data, 52000 types of 2 high-resolution imaging data, 36000 types of 3 high-resolution imaging data, and the test samples comprise 2000 types of 1 high-resolution imaging data, 2000 types of 2 high-resolution imaging data and 1200 types of 3 high-resolution imaging data.
3. Software environment for simulation experiment: the operating system is Windows 10 version, the processor is Intel (R) Core (TM) i5-7300HQ, and the main frequency of the processor is 2.50GHz; the software platform is as follows: python 3.5matlabr2016b, pytorch 0.4.
4. The simulation method comprises the following steps: the present invention and the existing methods, wherein the existing methods comprise the following 9 types:
a target recognition method based on a maximum correlation classifier,
a target recognition method based on an adaptive Gaussian classifier,
a target recognition method based on a linear support vector machine,
a target recognition method based on linear discriminant analysis and a support vector machine,
a target recognition method based on principal component analysis and combined with a support vector machine,
a target recognition method based on a deep belief network,
a target recognition method based on a stack noise reduction self-encoder,
a target recognition method based on a stack type related self-encoder combined with a support vector machine,
target identification method based on time domain convolutional neural network
(II) experimental contents and results:
experiment 1: the invention and the nine prior arts are adopted to respectively carry out target recognition on the high resolution range profile HRRP test samples of the 3 classes of aircrafts in the simulation condition, the ratio of the number of the test samples and the total number of the test samples, which are consistent with the class of the target recognition result and the class of the sample in each method, is respectively calculated, and the target recognition accuracy of each method is obtained, and the result is shown in table 1.
Table 1 target recognition accuracy list for ten methods
| Method | Recognition rate (%) |
| Maximum Correlation Classifier (MCC) | 62.42 |
| self-Adaptive Gaussian Classifier (AGC) | 85.63 |
| Linear Support Vector Machine (LSVM) | 86.70 |
| Linear discriminant analysis combined support vector machine (LDA) | 81.30 |
| Principal component analysis combined with support vector machine (PCA) | 83.81 |
| Deep Belief Network (DBN) | 89.29 |
| Stack noise reducing self-encoder (SDAE) | 90.42 |
| Stack type related self-encoder combined supportVector machine (SCAE) | 92.03 |
| Time domain convolutional neural network (TCNN) | 92.57 |
| The method of the invention | 93.01 |
As can be seen from table 1, in the ten methods of simulation experiment 1, the recognition accuracy of the method of the present invention is highest, which is 93.01%, and it is obvious that the target recognition performance of the method of the present invention is significantly better than that of the other nine methods. The correctness, the effectiveness and the reliability of the invention are verified by simulation experiments.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention; thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (5)
1. The HRRP target identification method based on the priori optimal variation self-encoder is characterized by comprising the following steps of:
(1) Extracting imaging data x of a radar echo of a Q-type target along a distance dimension on a radar line, and randomly dividing the Q-type high-resolution distance imaging data into a training sample set and a test sample set; wherein Q is a positive integer greater than 0;
(2) Carrying out mean normalization preprocessing on the imaging data to obtain preprocessed training set data x';
(3) Constructing a priori optimal variation self-coding machine composed of two perceptron,
(3a) Setting a first multi-layer perceptron which comprises two hidden layers, wherein each hidden layer is provided with 400 nodes, the input dimension of the perceptron is 256, and the output dimension is 60;
(3b) Setting a second multi-layer perceptron which comprises two hidden layers, wherein each hidden layer is provided with 400 nodes, the input dimension of the perceptron is 30, and the output dimension is 256;
(3c) Performing re-parametric processing on the output of the first multi-layer perceptron, and taking the output of the first multi-layer perceptron processed by the re-parametric processing as the input of the second multi-layer perceptron to form a priori optimal variation self-encoder;
(4) Training the priori optimal variation self-encoder by utilizing the preprocessed training set data x', and learning parameters of the first multi-layer perceptron and the second multi-layer perceptron to obtain a trained first multi-layer perceptron and a trained second multi-layer perceptron;
(5) Sequentially inputting training set data and test set data into a trained first multi-layer perceptron, and respectively extracting hidden variable features a corresponding to the training set and hidden variable features b corresponding to the test set;
(6) Inputting the hidden variable features a corresponding to the training set into a Support Vector Machine (SVM), training a support vector machine classifier, and classifying the hidden variable features b corresponding to the testing set by using the trained support vector machine classifier to obtain a target recognition result.
2. The method of claim 1, wherein the class Q high resolution range imaging data is randomly divided into a training sample set and a test sample set in (1) as follows:
(1a) Imaging data x of radar echoes along a distance dimension on a radar vision line are sequentially recorded as class 1 high-resolution distance imaging data, class 2 high-resolution distance imaging data, … and class Q high-resolution distance imaging data;
(1b) And extracting P training sample sets and A test sample sets from the Q-class high-resolution distance imaging data in a random extraction mode of the track segment, wherein each training sample or test sample contains N distance units to obtain a P multiplied by N-dimensional training sample set matrix and an A multiplied by N-dimensional test sample set matrix.
3. The method of claim 1, wherein the preprocessing of the imaging data in (2) for mean normalization is performed by the following formula:
where x' represents the preprocessed training set data, I 2 Representing a two-norm operation.
4. The method of claim 1, wherein (3 c) the output of the first multi-layer perceptron is re-parametrically processed as follows
(3c1) Equally dividing the output of the first multi-layer perceptron into two parts, respectively marking the two parts as mu and sigma;
(3c2) Sampling data from a gaussian distribution with a mean value of 0 and a variance of 1, and marking the data as epsilon;
(3c3) Calculating an input z of the second multi-layer perceptron from the results of (3 c 1) and (3 c 2):
z＝μ+ε×σ。
5. the method of claim 1, wherein (4) the a priori optimal variation self-encoder is trained using the preprocessed training set data x', as follows
(4a) Inputting the preprocessed training set data x' into a first multi-layer perceptron, and calculating the conditional log likelihood lnp corresponding to the x θ Lower variation bound of (x' |y):
wherein phi represents the parameter of the first multi-layer perceptron, theta 1 A priori distribution sufficiency statistic representative of the output of the first multi-layer perceptron, θ 2 Parameters of the second multi-layer perceptron, y represents the target class corresponding to x', p θ Representing a conditional distribution probability determined based on the second multi-layer perceptron output,
representing a variable lower bound function symbol;
(4b) The prior optimal variation is divided from the parameter phi of a first multi-layer perceptron of the encoder by using a batch random gradient descent method, and the prior distribution of the condition output by the first multi-layer perceptron is fully statistic theta 1 And a parameter θ of the second multilayer perceptron 2 And carrying out iterative updating 500 times through a maximum variation lower bound formula to obtain a trained first multi-layer perceptron and a trained second multi-layer perceptron.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910346042.3A CN110109110B (en) | 2019-04-26 | 2019-04-26 | HRRP target identification method based on priori optimal variation self-encoder |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910346042.3A CN110109110B (en) | 2019-04-26 | 2019-04-26 | HRRP target identification method based on priori optimal variation self-encoder |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110109110A CN110109110A (en) | 2019-08-09 |
| CN110109110B true CN110109110B (en) | 2023-06-16 |
Family
ID=67487007
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910346042.3A Active CN110109110B (en) | 2019-04-26 | 2019-04-26 | HRRP target identification method based on priori optimal variation self-encoder |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110109110B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111598163B (en) * | 2020-05-14 | 2022-02-01 | 中南大学 | Stacking integrated learning mode-based radar HRRP target identification method |
| CN111598881B (en) * | 2020-05-19 | 2022-07-12 | 西安电子科技大学 | Image anomaly detection method based on variational self-encoder |
| CN111580062B (en) * | 2020-05-25 | 2023-03-10 | 西安电子科技大学 | Target identification method based on double-circular-polarization high-resolution one-dimensional range profile HRRP |
| CN113239959B (en) * | 2021-04-09 | 2024-02-20 | 西安电子科技大学 | Radar HRRP target identification method based on decoupling characterization variation self-encoder |
| CN113361472B (en) * | 2021-07-01 | 2023-04-18 | 西安电子科技大学 | Radar HRRP target identification method based on ILFACs model |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108881196A (en) * | 2018-06-07 | 2018-11-23 | 中国民航大学 | The semi-supervised intrusion detection method of model is generated based on depth |
| CN109615577A (en) * | 2018-10-26 | 2019-04-12 | 西安电子科技大学 | High spectrum image super-resolution processing method based on convolutional network |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101617620B1 (en) * | 2014-09-30 | 2016-05-03 | 국방과학연구소 | Method and device for target classification by fusion of high resolution range profile and inverse synthetic aperture radar image |
| EP3612981B1 (en) * | 2017-04-19 | 2024-05-29 | Siemens Healthineers AG | Target detection in latent space |
| CN107609579B (en) * | 2017-08-25 | 2020-01-07 | 西安电子科技大学 | Radar target classification method based on steady variational self-encoder |
| CN107728143B (en) * | 2017-09-18 | 2021-01-19 | 西安电子科技大学 | Radar high-resolution range profile target identification method based on one-dimensional convolutional neural network |
| CN107977642B (en) * | 2017-12-15 | 2021-10-22 | 南京航空航天大学 | High-resolution range profile target identification method based on kernel self-adaptive mean discrimination analysis |
-
2019
- 2019-04-26 CN CN201910346042.3A patent/CN110109110B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108881196A (en) * | 2018-06-07 | 2018-11-23 | 中国民航大学 | The semi-supervised intrusion detection method of model is generated based on depth |
| CN109615577A (en) * | 2018-10-26 | 2019-04-12 | 西安电子科技大学 | High spectrum image super-resolution processing method based on convolutional network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110109110A (en) | 2019-08-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110109110B (en) | HRRP target identification method based on priori optimal variation self-encoder | |
| CN107728142B (en) | Radar high-resolution range profile target identification method based on two-dimensional convolutional network | |
| CN107728143B (en) | Radar high-resolution range profile target identification method based on one-dimensional convolutional neural network | |
| CN109376574B (en) | CNN-based (probabilistic neural network-based) HRRP (high-resolution Radar) target identification method for radar capable of refusing judgment | |
| CN110109109B (en) | HRRP target identification method based on multi-resolution attention convolution network | |
| CN111913156B (en) | Radar radiation source individual identification method based on deep learning model and feature combination | |
| CN108256436A (en) | A kind of radar HRRP target identification methods based on joint classification | |
| CN108764310B (en) | SAR target recognition method based on multi-scale multi-feature depth forest | |
| CN111458688B (en) | Three-dimensional convolution network-based radar high-resolution range profile target recognition method | |
| Wan et al. | Recognizing the HRRP by combining CNN and BiRNN with attention mechanism | |
| CN104732224B (en) | SAR target identification methods based on two-dimentional Zelnick moment characteristics rarefaction representation | |
| CN112882009A (en) | Radar micro Doppler target identification method based on amplitude and phase dual-channel network | |
| CN112137620B (en) | Ultra-wideband radar-based human body weak respiration signal detection method | |
| CN111401168A (en) | Multi-layer radar feature extraction and selection method for unmanned aerial vehicle | |
| CN114895263A (en) | Radar active interference signal identification method based on deep migration learning | |
| CN113239959B (en) | Radar HRRP target identification method based on decoupling characterization variation self-encoder | |
| CN111368653B (en) | Low-altitude small target detection method based on R-D graph and deep neural network | |
| CN115034257A (en) | Cross-modal information target identification method and device based on feature fusion | |
| CN113900101A (en) | Obstacle detection method and device and electronic equipment | |
| CN113486917A (en) | Radar HRRP small sample target identification method based on metric learning | |
| CN111914922A (en) | Hyperspectral image classification method based on local convolution and cavity convolution | |
| CN115061094B (en) | Radar target recognition method based on neural network and SVM | |
| CN116311067A (en) | Target comprehensive identification method, device and equipment based on high-dimensional characteristic map | |
| CN114137518A (en) | Radar high-resolution range profile open set identification method and device | |
| CN108106500A (en) | A kind of missile target kind identification method based on multisensor |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |