CN113449850A - Intelligent inhibition method for clutter of sea surface monitoring radar - Google Patents
Intelligent inhibition method for clutter of sea surface monitoring radar Download PDFInfo
- Publication number
- CN113449850A CN113449850A CN202110756212.2A CN202110756212A CN113449850A CN 113449850 A CN113449850 A CN 113449850A CN 202110756212 A CN202110756212 A CN 202110756212A CN 113449850 A CN113449850 A CN 113449850A
- Authority
- CN
- China
- Prior art keywords
- clutter
- data
- generator
- suppression
- clutter suppression
- 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
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012544 monitoring process Methods 0.000 title claims description 8
- 230000005764 inhibitory process Effects 0.000 title abstract description 4
- 230000001629 suppression Effects 0.000 claims abstract description 83
- 238000013507 mapping Methods 0.000 claims abstract description 19
- 238000012549 training Methods 0.000 claims description 25
- 230000006872 improvement Effects 0.000 claims description 10
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 claims description 5
- 238000011056 performance test Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 6
- 208000009119 Giant Axonal Neuropathy Diseases 0.000 abstract description 2
- 201000003382 giant axonal neuropathy 1 Diseases 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of 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
- G01S7/414—Discriminating targets with respect to background clutter
-
- 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
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention discloses an intelligent inhibition method for sea surface surveillance radar clutter, which is applied to the field of radar target detection and aims at solving the problem that the existing sea surface clutter inhibition technology is difficult to deal with complex and variable marine environments; the method comprises the steps of constructing two mirror symmetry generation countermeasure networks, wherein one path learns the mapping from original clutter data to data after clutter suppression, the other path learns the mapping from the data after clutter suppression to the original clutter data in reverse, two sets of data sets are respectively input into two paths of GANs, and the clutter suppression network with the sea clutter suppression function is finally obtained through the mutual constraint of two sets of generators and countermeasures and the dynamic identification of radar data before and after clutter suppression; the method of the invention can obviously improve the signal-to-clutter ratio of the target and completely reserve the original information of the target.
Description
Technical Field
The invention belongs to the field of radar target detection, and particularly relates to a radar sea clutter suppression technology.
Background
The sea surface monitoring radar is mainly used for detecting marine ships and tracking and measuring marine targets, and has important significance in military and civil use. In the process of detecting the sea by the sea surface monitoring radar, the echo includes not only a target echo but also a backscattered echo of the sea surface, which is also called a sea clutter. The sea clutter has a very complex dynamic characteristic, and when a sea surface target, especially a small target, is detected, the detection signal-to-clutter ratio of the target can be greatly reduced due to the existence of the sea clutter because the echo of the target is weak, so that the detection performance of a sea surface monitoring radar system is seriously influenced. Therefore, in order to improve the detection performance of the low detectable targets on the sea surface, sea clutter suppression studies must be conducted to eliminate or reduce the influence of sea clutter.
Currently, for the sea clutter suppression technology, the main method is to process radar echo signals in time domain, frequency domain, space-time combined domain, transform domain, and the like. The document "HF-over-the-radar shield detection with short dwells using cancellation. root, Benjamin T.RadioScience 33.4(1998): 1095-1111" proposes a sea clutter cyclic cancellation method which assumes that sea clutter satisfies a sinusoidal signal model, removes the strongest single frequency component at each iteration until the clutter is suppressed to a noise floor level, which requires a strong signal-to-noise ratio, and if the target and clutter Doppler frequencies are too close, it may cause the target to be suppressed. The document "an improved OTHR self-adaptive sea clutter suppression method, Zhao Shi Ching, Cheng Wen, Bao, System engineering and electronic technology, 2012,34(05): 909-.
Although the sea clutter suppression method can well suppress the sea clutter under specific conditions, the generalization capability of the sea clutter suppression method is weak, and the sea clutter suppression method is difficult to cope with complicated and variable marine environments.
Disclosure of Invention
In order to solve the technical problem, the invention provides an intelligent sea surface monitoring radar clutter suppression method, which realizes the suppression of sea clutter by designing two groups of mirror symmetric generation countermeasure Networks (GAN), learning the change characteristic of the sea clutter by using a generator and dynamically identifying the generator and an identifier in the generation countermeasure network.
The technical scheme adopted by the invention is as follows: a sea surface monitoring radar clutter intelligent suppression method comprises the following steps:
s1, acquiring sea surface radar images, including: original clutter data and data after clutter suppression;
s2, constructing two mirror symmetry generation countermeasure networks, wherein generators in two paths and a discriminator are mutually constrained;
and S3, inputting the original clutter data and the clutter suppressed data in the step S1 into two mirror symmetry generation countermeasure networks respectively for training, and finally obtaining the sea clutter intelligent suppression network.
Step S2 specifically includes:
two mirror symmetry generation countermeasure networks are constructed and are respectively marked as: the first path generates a countermeasure network, and the second path generates a countermeasure network; the first path generation countermeasure network comprises a first generator, a second generator and a first discriminator; the second path generation countermeasure network comprises a third generator, a fourth generator and a second discriminator;
the first generator is used for learning the mapping relation from the original clutter data to the data after clutter suppression; learning radar image characteristics of the data subjected to clutter suppression by using a second discriminator, and discriminating an output result of the first generator; the second generator is used for learning the mapping relation from the data subjected to clutter suppression to the original clutter data;
the third generator is used for learning the mapping relation from the data after clutter suppression to the original clutter data; learning radar image characteristics of the original clutter data by using a first discriminator, and identifying an output result of a third generator; the fourth generator is used for learning the mapping relation of the original clutter data to the data after clutter suppression.
The first generator, the second generator, the third generator and the fourth generator have the same structure, and each of the four generators comprises: the radar image mapping method comprises a convolutional layer, a residual block and an deconvolution layer, wherein the convolutional layer and the residual block are used for extracting deep features of an input radar image, and the deconvolution layer is used for reconstructing the radar image to obtain a mapping result.
Still include step S3 the sea clutter intelligent suppression network carry out performance test, it is specific: measuring clutter background improvement levels of the radar image before and after clutter suppression by calculating a sea clutter improvement factor; the calculation formula of the sea clutter improvement factor is as follows:
σ＝SNRout-SNRin
wherein the SNRout、SNRinAnd respectively representing the signal-to-noise ratio of the radar image after clutter suppression and before clutter suppression, wherein the unit is dB.
Step S3, the training comprises two stages, wherein in the first stage, a single-target high signal-to-noise ratio data set is adopted for training to obtain a preliminary sea clutter suppression network; and in the second stage, on the basis of the primary sea clutter suppression network obtained in the first stage, training is carried out by adopting a data set with the number of the targets increased, so as to obtain a final sea clutter suppression network.
The invention has the beneficial effects that: compared with the prior art, the method is based on the machine learning correlation theory, has stable sea clutter suppression performance, can remarkably improve the signal-to-clutter ratio of the target and completely reserve the original information of the target, and can adapt to different ocean scenes.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention;
FIG. 2 is a sea clutter suppression network framework of the method of the present invention;
FIG. 3 is a schematic structural diagram of a sea clutter suppression network generator according to the method of the present invention;
FIG. 4 is a schematic diagram of a sea clutter suppression network discriminator according to the method of the present invention;
fig. 5 shows the clutter suppression result of the simulation data according to this embodiment:
fig. 5(a) shows clutter suppression results of different numbers of targets, and fig. 5(b) shows clutter suppression results of different input signal-to-clutter ratios;
fig. 6 is a clutter suppression result on the measured data according to the present embodiment;
fig. 6(a) is a radar image before clutter suppression, fig. 6(b) is a radar image after clutter suppression, fig. 6(c) is an amplitude distribution diagram corresponding to fig. 6(a), and fig. 6(d) is an amplitude distribution diagram corresponding to fig. 6 (b).
Detailed Description
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.
The process flow diagram of the method of the invention is shown in figure 1, and comprises the following steps:
A. acquiring a sea surface radar image;
the method comprises the following steps of constructing a data set based on a composite K distribution model to generate sea clutter radar echo data, dividing the data set into a clutter suppression front image set X and a clutter suppression back image set Y, and designing partial parameters of a radar set by a sea clutter model as shown in a table 1:
TABLE 1 Radar parameter settings
B. Designing a clutter intelligent suppression network;
the invention provides a novel sea clutter suppression idea, and two paths of mirror symmetric GANs are used for training a training set X and a training set Y generated in A and A in a non-paired mode. The influence of individual data on the network performance can be effectively reduced by a non-pairing training mode.
B1, clutter suppression network framework;
the frame of the sea clutter suppression network is shown in FIG. 2, wherein F and G denote generators, DXAnd DYThe discriminator is shown.
The invention uses a generator F to learn the mapping, and X (X belongs to X) is converted by F to obtain F (X) which is the clutter suppression result of X. To ensure F (x) ε Y, a discriminator D is usedYAnd learning the radar image characteristics of Y, and identifying F (X), and guiding F to learn the mapping relation from X to Y, wherein the process can be seen in the left branch in FIG. 2.
In order to ensure that F (X) has a one-to-one correspondence with X and avoid the situation that F (X) has changed target information although having the image characteristics of Y space, the invention uses the generator G to convert Y (Y ∈ Y) into the picture G (Y) in X. To ensure G (y) ε X, a discriminator D is usedXAnd learning the radar image characteristics of X, identifying G (Y), and guiding G to learn the mapping relation from Y to X, wherein the process can be seen in the right branch in FIG. 2. Reducing clutter suppression result F (x) by mapping G, and recording
Then
By the arrangement, the one-to-one correspondence of x and F (x) can be realized, and the clutter suppression effect is ensured. The optimization process for F is shown on the left side and the optimization process for G is shown on the right side of fig. 2.
B2 structural design of generator and discriminator
The network structures and parameter settings of the generator and the discriminator are respectively shown in fig. 3 and fig. 4, the generator extracts deep features of the input radar image by using the convolution layer and the residual block, improves the problem of network degradation, improves training efficiency, and reconstructs the radar image by using the deconvolution layer to obtain a mapping result. The discriminator uses the multi-layer convolutional layer to distinguish the original image from the image resulting from the generator reconstruction.
B3 loss function design
According to the design ideas of B1 and B2, the designed loss function is as follows:
wherein L isGAN(F,DY,X,Y)、LGAN(G,DX,Y,X)、Lcyc(G,F)、LmatThe expression (F, G) is as follows:
wherein λ represents a hyper-parameter, λ is learned automatically by a training process, F and G are generators, DXAnd DYFor the discriminator, X and Y are images in X and Y, F (X) and F (Y) corresponding to the X and Y reconstructed images, Pdata(x) And Pdata(y) is the data distribution of the real image,
and
is an expectation function of the distribution of real image data, | · | | luminance1Represents L1And (4) norm. Has been optimizedThe program can be expressed as
The formulas (2) and (3) are standard optimization terms of two-way mirror image GAN, and the information such as the structure of the input image is reserved by using the formula (4).
Introducing matching losses Lmat(F, G): and adding an excitation item Y (Y belongs to Y) → F (Y) (F (Y) Y) and X (X) → G (X) (G (X)) X to G and F, ensuring the stability of the target area in the clutter suppression process, realizing the pixel level correspondence between the image after the clutter suppression and the image before the clutter suppression, and effectively retaining the information of the target area.
C. Network training
The characteristics of the sea clutter are very complicated, and in order to enable the network to better learn the corresponding characteristics of the sea clutter, a pre-training mode is adopted in training, so that the performance of the sea clutter suppression network is gradually improved. In the pre-training stage, a single-target high signal-to-noise ratio data set is used for pre-training to obtain a sea clutter suppression network, and in the embodiment, the training data set in the first stage is 2000 pieces. And in the second stage, the number of targets of the training data set is increased appropriately, and 1 to 3 targets are added randomly. Benefiting from the pre-training, the second stage can reduce the training data, the training data set is set to be 500-1000, meanwhile, the pre-training greatly shortens the training period required by adjusting the hyper-parameters, and the rapid parameter adjustment can be realized.
D. Performance testing
In the testing stage, the invention tests aiming at different situations to evaluate the clutter suppression performance of the proposed sea clutter suppression model, introduces a sea clutter improvement factor, and is defined as follows:
σ＝SNRout-SNRin (6)
wherein the SNRout、SNRinAnd respectively representing the signal-to-noise ratio of the radar image after clutter suppression and before clutter suppression, wherein the unit is dB. And measuring the clutter background improvement level of the radar image before and after clutter suppression by comparing the sea clutter improvement factors, thereby verifying the clutter suppression performance of the sea clutter suppression method.
The following are the results of the performance test experiments
Fig. 5 is a clutter suppression result on simulation data according to this embodiment, where fig. 5(a) lists suppression results of models for different numbers of targets, and fig. 5(b) shows clutter suppression results of models for different input signal-to-clutter ratios. It can be seen from the figure that the model has strong robustness under the condition that the target number position and the input signal-to-noise ratio are uncertain. Fig. 6 shows clutter suppression results on actual measurement data according to the present embodiment. As can be seen from the figure, the clutter level is significantly reduced and the target is left intact. The intelligent clutter suppression network obtained by the invention has strong robustness under the conditions of uncertain target quantity and position and input signal-to-clutter ratio.
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 (5)
1. An intelligent sea surface monitoring radar clutter suppression method is characterized by comprising the following steps:
s1, acquiring sea surface radar images, including: original clutter data and data after clutter suppression;
s2, constructing two mirror symmetry generation countermeasure networks, wherein generators in two paths and a discriminator are mutually constrained;
and S3, inputting the original clutter data and the clutter suppressed data in the step S1 into two mirror symmetry generation countermeasure networks respectively for training, and finally obtaining the sea clutter intelligent suppression network.
2. The method for intelligently suppressing clutter of a sea surface surveillance radar according to claim 1, wherein the step S2 specifically comprises:
two mirror symmetry generation countermeasure networks are constructed and are respectively marked as: the first path generates a countermeasure network, and the second path generates a countermeasure network; the first path generation countermeasure network comprises a first generator, a second generator and a first discriminator; the second path generation countermeasure network comprises a third generator, a fourth generator and a second discriminator;
the first generator is used for learning the mapping relation from the original clutter data to the data after clutter suppression; learning radar image characteristics of the data subjected to clutter suppression by using a second discriminator, and discriminating an output result of the first generator; the second generator is used for learning the mapping relation from the data subjected to clutter suppression to the original clutter data;
the third generator is used for learning the mapping relation from the data after clutter suppression to the original clutter data; learning radar image characteristics of the original clutter data by using a first discriminator, and identifying an output result of a third generator; the fourth generator is used for learning the mapping relation of the original clutter data to the data after clutter suppression.
3. The intelligent sea surface surveillance radar clutter suppression method according to claim 2, wherein the first generator, the second generator, the third generator and the fourth generator have the same structure, and each of the four generators comprises: the radar image mapping method comprises a convolutional layer, a residual block and an deconvolution layer, wherein the convolutional layer and the residual block are used for extracting deep features of an input radar image, and the deconvolution layer is used for reconstructing the radar image to obtain a mapping result.
4. The intelligent sea surface surveillance radar clutter suppression method according to claim 1, further comprising performing a performance test on the intelligent sea clutter suppression network of step S3, specifically: measuring clutter background improvement levels of the radar image before and after clutter suppression by calculating a sea clutter improvement factor; the calculation formula of the sea clutter improvement factor is as follows:
σ＝SNRout-SNRin
wherein the SNRout、SNRinRespectively representing the signal-to-noise ratio of the radar image after clutter suppression and before clutter suppression,the unit is dB.
5. The intelligent sea surface surveillance radar clutter suppression method according to claim 1, wherein the training of step S3 includes two phases, the first phase is trained with a single-target, high signal-to-clutter ratio data set to obtain a preliminary sea clutter suppression network; and in the second stage, on the basis of the primary sea clutter suppression network obtained in the first stage, training is carried out by adopting a data set with the number of the targets increased, so as to obtain a final sea clutter suppression network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110756212.2A CN113449850A (en) | 2021-07-05 | 2021-07-05 | Intelligent inhibition method for clutter of sea surface monitoring radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110756212.2A CN113449850A (en) | 2021-07-05 | 2021-07-05 | Intelligent inhibition method for clutter of sea surface monitoring radar |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113449850A true CN113449850A (en) | 2021-09-28 |
Family
ID=77815135
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110756212.2A Pending CN113449850A (en) | 2021-07-05 | 2021-07-05 | Intelligent inhibition method for clutter of sea surface monitoring radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113449850A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117217103A (en) * | 2023-11-09 | 2023-12-12 | 南京航空航天大学 | Satellite-borne SAR sea clutter generation method and system based on multi-scale attention mechanism |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102176000A (en) * | 2011-01-27 | 2011-09-07 | 电子科技大学 | Sea clutter suppression method for marine radar |
| CN108875935A (en) * | 2018-06-11 | 2018-11-23 | 兰州理工大学 | Based on the natural image target materials visual signature mapping method for generating confrontation network |
| US20190026867A1 (en) * | 2017-07-21 | 2019-01-24 | The Boeing Company | Suppression of background clutter in video imagery |
| CN111242064A (en) * | 2020-01-17 | 2020-06-05 | 山东师范大学 | Pedestrian re-identification method and system based on camera style migration and single marking |
| CN111580099A (en) * | 2020-06-12 | 2020-08-25 | 南京信息工程大学 | Wall clutter suppression method of through-wall imaging radar based on joint entropy |
| CN111626961A (en) * | 2020-05-29 | 2020-09-04 | 中国人民解放军海军航空大学 | Radar image clutter suppression method and system based on generation countermeasure network |
| CN111627429A (en) * | 2020-05-20 | 2020-09-04 | 浙江工业大学 | Defense method and device of voice recognition model based on cycleGAN |
| CN111723840A (en) * | 2020-05-08 | 2020-09-29 | 天津大学 | Clustering and style migration method for ultrasonic images |
| CN111986108A (en) * | 2020-08-07 | 2020-11-24 | 西北工业大学 | Complex sea-air scene image defogging method based on generation countermeasure network |
-
2021
- 2021-07-05 CN CN202110756212.2A patent/CN113449850A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102176000A (en) * | 2011-01-27 | 2011-09-07 | 电子科技大学 | Sea clutter suppression method for marine radar |
| US20190026867A1 (en) * | 2017-07-21 | 2019-01-24 | The Boeing Company | Suppression of background clutter in video imagery |
| CN108875935A (en) * | 2018-06-11 | 2018-11-23 | 兰州理工大学 | Based on the natural image target materials visual signature mapping method for generating confrontation network |
| CN111242064A (en) * | 2020-01-17 | 2020-06-05 | 山东师范大学 | Pedestrian re-identification method and system based on camera style migration and single marking |
| CN111723840A (en) * | 2020-05-08 | 2020-09-29 | 天津大学 | Clustering and style migration method for ultrasonic images |
| CN111627429A (en) * | 2020-05-20 | 2020-09-04 | 浙江工业大学 | Defense method and device of voice recognition model based on cycleGAN |
| CN111626961A (en) * | 2020-05-29 | 2020-09-04 | 中国人民解放军海军航空大学 | Radar image clutter suppression method and system based on generation countermeasure network |
| CN111580099A (en) * | 2020-06-12 | 2020-08-25 | 南京信息工程大学 | Wall clutter suppression method of through-wall imaging radar based on joint entropy |
| CN111986108A (en) * | 2020-08-07 | 2020-11-24 | 西北工业大学 | Complex sea-air scene image defogging method based on generation countermeasure network |
Non-Patent Citations (3)
| Title |
|---|
| MINGLYU 等: "Clutter suppression method for linear scan indoor near-field RCS measurement based on interferometric SAR", 《PROCEDIA COMPUTER SCIENCE》 * |
| 张文武: "基于生成式对抗网络的声呐和雷达图像增强方法", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
| 李山川: "双基前视SAR非平稳杂波抑制方法研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117217103A (en) * | 2023-11-09 | 2023-12-12 | 南京航空航天大学 | Satellite-borne SAR sea clutter generation method and system based on multi-scale attention mechanism |
| CN117217103B (en) * | 2023-11-09 | 2024-03-15 | 南京航空航天大学 | Satellite-borne SAR sea clutter generation method and system based on multi-scale attention mechanism |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109557429A (en) | Based on the GIS partial discharge fault detection method for improving wavelet threshold denoising | |
| CN110221256A (en) | SAR disturbance restraining method based on depth residual error network | |
| CN113449850A (en) | Intelligent inhibition method for clutter of sea surface monitoring radar | |
| CN107561540A (en) | The detection method and device of sonar leakage target based on the symmetrical spectral property of reverberation | |
| CN112329764A (en) | Infrared dim target detection method based on TV-L1 model | |
| Wang et al. | Polarization image fusion algorithm using NSCT and CNN | |
| CN111090089B (en) | Space-time adaptive detection method based on two types of auxiliary data | |
| Wang et al. | Seismic multiple suppression based on a deep neural network method for marine data | |
| CN107479050A (en) | Object detection method and device based on symmetrical spectral property and sub-symmetry characteristic | |
| CN114943245A (en) | Automatic modulation recognition method and device based on data enhancement and feature embedding | |
| Yi et al. | A bearing fault diagnosis method based on an improved depth residual network | |
| CN112505665B (en) | Space-time self-adaptive detection method and system suitable for partial uniform reverberation environment | |
| CN114117912A (en) | Sea clutter modeling and inhibiting method under data model dual drive | |
| CN111090088B (en) | Leakage target space-time detection method based on active sonar array skew symmetry characteristics | |
| CN115356599B (en) | Multi-mode urban power grid fault diagnosis method and system | |
| CN116106829A (en) | Intelligent target detection method under partial uniform clutter and interference | |
| CN105675983B (en) | The detection of weak harmonic signal and method for reconstructing under a kind of strong Chaotic Background | |
| Mao et al. | DenoiseNet: Deep generator and discriminator learning network with self-attention applied to ocean data | |
| Changbo et al. | Radar signal separation and recognition based on semantic segmentation | |
| Wu et al. | A machine learning approach to clutter suppression for marine surveillance radar | |
| CN112926534A (en) | SAR (synthetic aperture radar) graphic ship target detection method based on transform domain information fusion | |
| CN113516187A (en) | Infrared weak and small target detection algorithm adopting local characteristic contrast | |
| Han et al. | A Bootstrap-Bayesian dynamic modification model based on small sample target features | |
| CN115685081B (en) | GLRT-based method for detecting distance expansion target in interference plus noise background | |
| Zhan et al. | Tamura coarseness for evaluating OTH radar image and its application in RFI suppression |
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: 20210928 |
|
| RJ01 | Rejection of invention patent application after publication |