Search

Compared to visual sensors that have light dependence and privacy intrusion issues, non-intrusion millimeter-wave (mmw) radars are more suitable for the daily person identification. In a realistic home scenario, there are new challenges that are not taken into account in existing research. This paper attempts to address these issues such as multipath interference, complex walking process and recognition robustness in smart home scenarios, and designs a lightweight multi-branch convolutional neural network with an Inception-Pool module and a Residual-Pool module to learn and classify gait Doppler features. The experimental results in a home living room scenario indicate that the designed mmw radar person identification system can achieve accurate and robust real-time identification performance.

Ship detection using polarimetric synthetic radar (PolSAR) imagery attracts a lot of attention in recent years. Most notably, the detector polarimetric notch filter (PNF) has been demonstrated to be effective for ship detection in PolSAR imagery, which gives excellent performances. In this work, a mathematical form of one new PNF (NPNF) based on physical mechanisms of targets and clutter is further developed for partial targets. The different mechanisms have been revealed based on the projection matrix. The experimental results including simulated and measured data demonstrate that the NPNF exhibits a better performance than the original PNF.

Human motion recognition based on ultra-wideband through-the-wall radar (UWB TWR) (a radar whose fractional bandwidth of the radar transmitted signal is bigger than 0.25) is faced with the problems of too few samples and the limitation of perspective. In this letter, we propose a multiradar cooperative human motion recognition model based on transfer learning and ensemble learning. Specifically, a ResNeXt network model based on transfer learning is first proposed to deal with the problem of too few samples. The model is pretrained on the public ImageNet database, and then it is transferred to the task of human motion recognition based on multiradar. Compared with a typical convolutional neural network from scratch, the ResNeXt network model based on transfer learning requires shorter epochs and achieves higher accuracy. Then, to solve the problem of model accuracy decline caused by the limitation of perspective, a multiradar human motion recognition model based on ensemble learning is proposed. Experimental results show that compared with the fusion model based on single-view radar, the recognition accuracy of network based on ensemble learning can be higher.

Protective measures against small unmanned aerial vehicles (UAVs) are vital from a national security perspective. As a result, the importance of surveillance systems that automatically identify and classify low radar cross section (RCS) aerial targets increases. In this work, an indigenously developed continuous wave (CW) (X-band: 10 GHz) radar is used to build a diversified ``DIAT-μSAT'' dataset comprising 4849 micro-Doppler signature images of five different small aerial targets. We also proposed a transfer learning-based deep convolutional neural network (DCNN) approach for classifying low RCS aerial targets. We demonstrated the classification accuracy of 95% and 97%, with VGG16 and VGG19 as feature extractors, respectively, with minimal false-negative and -positive results. The open-field experimental classes covered in this work are: 1) a two-blade rotor; 2) a three-short-blade rotor; 3) a three-long-blade rotor; 4) a quadcopter; 5) a bionic bird; and 6) a two-blade-rotor and bionic bird. We also observed a good classification accuracy (>97%) when more than one target is operated simultaneously.

Existing works on radar high-resolution range profile (HRRP) recognition commonly focus on utilizing data and various deep learning models in achieving high classification accuracy. However, in practical applications, it is often difficult to obtain HRRP signals, especially for noncooperative targets. Such lack of data dramatically decreases the recognition performance, so this letter applies data augmentation to address small-sample problems. A recognition-aware HRRP generation framework based on a generative adversarial network is proposed for data augmentation, which generates discriminative samples by decomposing and reorganizing signal's characteristics. The proposed model increases the generated signals' discriminative power, thus meeting the application requirements. Experiments show that the generated HRRP signals can not only accurately expand the data set but also improve the recognition system's performance. Besides, the developed model outperforms traditional data augmentation methods and other generative methods. To the best of our knowledge, this is the first work on HRRP signal generation in radar automatic target recognition systems.

A cascaded regional people counting approach based on two-dimensional (2-D) spatial attribute features using multi-input multi-output (MIMO) radar is presented in this letter. The main task is to detect whether there is a target, and to classify the number of people in a closed metal ramp. The difficulty to accurately detect the number of people lies in the presence of strong static clutter and the strong multipath effects in the environment. In this letter, range-angle spectrum is first computed and exploited to extract the 2-D spatial attribute features, namely the effective peaks of range-angle spectrum, the spacing features and the energy block features. These features reflect the relative position relationships between the people and the environment as well as the physical structure of a certain environment. Then, a maximum likelihood classifier and a random forest classifier are cascaded and applied to detect the target presence and classify the number of people. The experiment results show that the average classification accuracy of 0 person, 1 person, and multiple persons is above 92.55%, which outperforms the existing people counting approaches.

Insufficient data of a noncooperative target seriously affect the performance of radar automatic target recognition (RATR) using the high-resolution range profile (HRRP), especially when the noncooperative target has only one sample. To this end, we propose an unsupervised data generation method to generate noncooperative HRRP signals. We utilize the pretrained generative adversarial networks (GANs) model to learn the HRRP general probability distribution. To emphasize the representative and discriminative power of generated HRRP signals, a joint optimization method is proposed to preserve category information. Moreover, a feature diversification method is proposed to make the generated samples have sufficient aspect characteristics to further fit the probability distribution of the noncooperative target. Thus, the generated HRRP signals can effectively improve the recognition performance of noncooperative target. Extensive experiments on HRRP data sets demonstrate the superior performance of our method over other state-of-the-art methods.

The holographic subsurface radar (HSR) is recognized as an effective remote sensing modality for the detection of shallowly buried objects with a high-resolution image in plain view. However, subsurface detection with HSR is prone to be impaired by clutter contamination, which often obscures the target response. In this letter, a novel clutter mitigation method combining singular value decomposition (SVD) and response cross correlation analysis is presented. The proposed method first applies SVD to decompose the radar data matrix to a number of singular components. Furthermore, the signal cross correlation characteristics are analyzed to demonstrate that the variance of left singular vectors is directly proportional to the target proportion in radar data. Then, target and clutter subspaces can be identified by maximizing the defined weighted target-to-clutter ratio (WTCR). Results of numerical simulation and laboratory experiments corroborate the effectiveness of the proposed method in reducing clutter while preserving the target image.

Classification of nonprecipitation echoes of radar is an inevitable step in radar-based precipitation estimation. Among nonprecipitation echoes, clear air echoes are specifically difficult to distinguish for their similarity to precipitation echoes. This letter aims to conduct a pixelwise classification of clear air echoes for image sequences of the radar reflectivity. We propose the Static and Motion streams Network (SMNet) to simultaneously utilize the static and motion features. SMNet realizes capturing the spatiotemporal characteristics while maintaining the details of the current frame via a fusion structure and a novel training method. For feature fusion, the static and motion streams are concatenated. Then, for model training, we adopt a dynamic weight assignment strategy to further extract rich information. Finally, we validate our method on an S-band single-polarization radar in Beijing, China, from May to September 2018. The results demonstrate that the overall performance of SMNet is superior to other competitors.

Convolutional neural network (CNN)-based methods have become the mainstream in radar ship recognition. However, these methods suffer from two common problems. First, the training samples consist largely of common ship types, giving them an overwhelming numerical advantage over rare ship types. As a result, CNN-based recognition algorithms fail to classify rare ship types correctly. Second, huge high-resolution slices result in heavy computational burdens. To solve the first problem, namely, the class imbalance problem, this letter proposes a CNN training method that combines deep metric learning (DML) with gradually balanced sampling. DML obtains the center of each class in the feature space and performs clustering equally. Gradually balanced sampling adopts a smooth transition from instance-aware resampling to class-aware resampling to improve the recognition rate drop caused by traditional resampling methods. As for the second problem, to reduce the computational complexity of high-resolution synthetic aperture radar (SAR) images, a lightweight CNN is also proposed.

Non-stationary moving target detection in spiky sea clutter is a challenging task due to the high-power, time-varying, and target-like properties of sea spikes. In this paper, we propose a time-frequency correlation (TFC)-based constant false alarm rate (TFC-CFAR) detection method on the Time-frequency manifold, and apply to the non-stationary moving target detection in spiky sea clutter. The data samples in each range cell are modeled as a TFC matrix that captures the correlation between two frequency components of the time frequency distribution. The clutter covariance matrix is estimated by the geometric mean of a set of TFC matrices in reference cells. Three geometric metrics are employed to measure the dissimilarity between the clutter and target signals. Based on these geometric measures, three TFC-CFAR detectors are compared. Experiments performed on real IPIX radar dataset confirm that the TFC can be used for identifying and eliminating sea spikes, while the TFC-CFAR detector achieves better detection performance than the conventional detectors.

High-frequency surface wave radar (HFSWR) can be effectively used to detect ships in the exclusive economic zone. However, the ship signal is concealed and interfered with various clutter and background noise in the Doppler spectrum. In this letter, a range-Doppler (RD) image-based novel ship detection algorithm is proposed by exploiting spatial-frequency information and a unique feature fusion based on the analysis of variance. The algorithm subsumes three successive stages: Stage I--the plausible region of interest is captured, Stage II--the features from different sources are fused into one generalized feature space, and Stage III--an extreme learning machine-based classifier is utilized to localize the ships. Experimental results on challenging HFSWR-RD datasets demonstrate that the proposed algorithm has a competitive performance over other ship detection algorithms.

With the development of the radar sensing technology, hand gesture sensing and recognition has attracted much attention. This letter adopts a frequency-modulated continuous wave (FMCW) radar to achieve short-range hand gesture sensing and recognition. Specifically, the range, Doppler, and angle parameters of hand gestures are measured by fast Fourier transformation (FFT) and multiple signal classification (MUSIC) algorithm, respectively. The mixup (MP) algorithm combined with augmentation (AU) algorithm using a weight factor is applied to expand the hand gesture data. Then, a complementary multidimensional feature fusion network-based hand gesture recognition (CMFF-HGR) is designed to extract the features and achieve HGR. Finally, a series of experiments are carried out to verify the effectiveness of the proposed approach, and the results show that the recognition accuracy is higher than the existing alternatives with low computational complexity.

The higher bandwidth inverse synthetic aperture radar (ISAR) can obtain the higher resolution radar images, which can provide more target information and help improve radar target detection and recognition. It is essential to study how to achieve a precise high-resolution (HR) ISAR image utilizing limited measurement echoes. The existing neural-network-based ISAR imaging methods extract features only from limited measurement echoes, and the common features in HR ISAR images are not utilized sufficiently, which limits the imaging performance improvement. Moreover, in their loss functions, there are no explicit constraints on the correct recovery of strong scattering points, which are important in reflecting the target characteristics. In this letter, we propose a conditional probabilistic generative model to achieve the HR ISAR imaging. By optimizing the well-designed Kullback-Leibler (KL) divergence between conditional prior and approximate posterior probability distribution in the loss function, the common features contained in training HR radar images can be learned, and a suitable prior probability distribution for the latent variable can be obtained. To accurately recover the positions and relative amplitudes of strong scattering points, we blend a similarity measurement that is sensitive to the large values' locations in a vector with the adversarial loss. Both visual and numerical results of extensive experiments prove that the proposed model can obtain enhanced effectiveness and efficiency compared with some counterparts.

Ground-Penetrating Radar (GPR) technology plays an important role in near-surface geophysical survey. Usually, the dataset acquired by higher frequency antenna has higher resolution but limited penetration depth; while the dataset collected by lower one has deeper penetration depth but lower resolution. It is necessary to fuse several different frequency datasets into a composite displayed profile to get a trade-off between penetration depth and resolution. In such case, we propose a novel multi-frequency GPR data fusion method and test it on the data acquired from a limestone quarry in the Taihang Mountains of China. In our algorithm, time-varying weighting factors are calculated by virtues of a sliding time window, i.e. designed with segmented signals. According to this approach, we can achieve better local characteristics of the signal as well as the smoothness of the signal waveform. The results show that the spectrum bandwidth of the fused signal is broadened, and the fusion profile significantly improves the imaging of internal structures for limestone on different scales.

Different from conventional radar systems, for the single-photon radar, a single-photon detector acts as the receiver to detect the echo signal with only a few photons, which can be used to achieve ultrahigh sensitivity for long-distance target detection. In this letter, the heterodyne single-photon radar at terahertz frequencies is studied, for the first time. First, the working principle of the developed heterodyne single-photon radar is presented, and the target echo signal is derived. Subsequently, the estimation method of the target distance and the Doppler frequency is proposed, and the performance is analyzed by probability density function (PDF) derivation. Simulation results show good agreement with the theoretical analysis. This work can pave the way to the development of a new radar paradigm.

Ultrawideband (UWB) multiple-input-multiple-output (MIMO) radar is widely used for through-wall imaging (TWI) due to its excellent penetrability and large aperture. Multichannels in the MIMO radar system are usually time-division multiplexing based on microwave switches to reduce the complexity of the system in engineering. The switching process of the channel will bring time delay, which cannot be ignored in the TWI of the moving target. The switching time delay will cause the defocus and position shift of the TWI of the moving target. This letter proposes a motion compensation method based on multiframe data fusion (MFDF) used for correcting the echo of the through-wall moving target. A geometric model is established in the proposed method through the echo of the current frame and the next frame, and the compensated signal is obtained through the geometric solution. The proposed method is compared with before compensation and the traditional single-channel motion compensation algorithm (SCMCA) through simulation and experimental data verification. The visual images and quantitative results show that the proposed motion compensation method can obtain a good focus image of the through-wall moving target and reduce the positioning error.

Ground-penetrating radar (GPR) is a well-known geophysical electromagnetic method used to detect the underground facilities such as landmines, pipelines, and cavities. In general, the clutter presented in GPR B-scan image obscures the underground objects, thus damaging the performance of the underground object detection algorithm. In this letter, we proposed a new clutter suppression method based on robust autoencoder (RAE). The proposed algorithm decomposes a GPR B-scan image into its low-rank and sparse components. The low-rank component catches the clutter, whereas the sparse component captures the underground object responses. The commonly used clutter removal algorithms, mean subtraction (MS), singular value decomposition (SVD), robust principal component analysis (RPCA), and morphological component analysis (MCA), are compared with the proposed algorithm on both the numerical simulated data and real GPR data. The visual and quantitative results demonstrate the effectiveness of the proposed RAE-based algorithm over the widely used state-of-the-art clutter removal algorithms.

Aerosol extinction coefficient (AEC) is a crucial factor of aerosol optical properties that have an important influence on the atmospheric process. As a combination of laser technology and radar technology, LiDAR is a powerful remote sensing technology for aerosol measurement. However, many uncertainties still exist in LiDAR data retrieval because frequently used retrieval methods need some assumptions or complex numerical operations for retrieving AEC, which may cause large deviation. In this letter, combining the advantages and disadvantages of Mie scattering LiDAR and high spectral resolution LiDAR (HSRL), a new inversion model based on deep belief network (DBN) is proposed to predict AEC, which can effectively avoid the uncertainty caused by many assumptions and improve the detection accuracy of Mie scattering LiDAR. The experiment results indicate that the trained DBN model is robust and satisfactory and provides a competitive solution for predicting AEC of Mie scattering LiDAR. In addition, this letter reveals that deep learning method has huge potential for LiDAR data retrieval.

With the development of the Internet of things, radar-based human activity recognition is becoming more and more important, because they play an indispensable role in fields such as safety and health monitoring. In this work, a novel network named 1-D dense attention neural network (1-D-DAN) is proposed for the radar-based human activity recognition. In the proposed network, a novel attention mechanism network structure specifically designed for radar spectrogram is proposed, equipping 1-D convolutional network with attention mechanism. With the x-axis of the spectrogram represents time and the y-axis represents frequency, the proposed attention mechanism includes two branches: 1) time attention branch and 2) frequency attention branch. Moreover, a dense attention operation that can make full use of features in the network is also introduced in the proposed attention mechanism. Experimental results show that compared with the state-of-the-art methods, our proposed 1-D-DAN achieves the highest accuracy in human activity recognition with the lowest computational complexity.

Low imaging spatial resolution hinders through-the-wall radar imaging (TWRI) from reconstructing complete human postures. This letter mainly discusses a convolutional neural network (CNN)-based human posture reconstruction method for TWRI. The training process follows a supervision-prediction learning pipeline inspired by the cross-modal learning technique. Specifically, optical images and TWRI signals are collected simultaneously using a self-develop radar containing an optical camera. Then, the optical images are processed with a computer-vision-based supervision network to generate ground-truth human skeletons. Next, the same type of skeleton is predicted from corresponding TWRI signals using a prediction network. After training, the model shows complete predictions in wall-occlusive scenarios solely using TWRI signals. Experiments show comparable quantitative results with the state-of-the-art vision-based methods in nonwall-occlusive scenarios and accurate qualitative results with wall occlusion.

Target detection in a high-frequency surface-wave radar (HFSWR) system is a challenging task because radar returns are strongly polluted by various sources of interference. To address the detection problem, this letter presents an automatic detection method based on multidirection dictionary learning for HFSWR. First, we perform clutter identification and statistical analysis to become aware of the time-varying clutter environment. The analysis of real data shows that the clutter in an HFSWR system has spatial correlations and geometric directions. Second, motivated by this information, we design a multidirection dictionary learning-based constant false alarm rate (MDDL-CFAR) detector in which the spatial and geometric direction information is well represented by multidirectional dictionaries. The MDDL-CFAR can simultaneously learn dictionaries and estimate clutter statistics to adaptively set the detection threshold. Experimental results on HFSWR data sets demonstrate the effectiveness of the proposed detection method.

This letter deals with an imaging strategy able to manage effectively data collected on multiple lines by means of a Mini-Unmanned Aerial Vehicle (M-UAV) radar system operating in Sounder mode. The strategy, named multilines imaging approach (MIA), allows an effective 3-D pseudo-representation of the investigated volume. At the first step, MIA faces the problem of reconstructing 2-D domains (slices) by exploiting data collected on one or more lines. Then, MIA interpolates the 2-D reconstructions to provide the 3-D representation. The imaging of each slice is formulated as a linear inverse scattering problem, which is solved by means of the truncated singular value decomposition (TSVD) regularization scheme. The MIA effectiveness is assessed by processing real data collected at Archeological Park of Paestum and Velia, Paestum, Italy.

Radio frequency interference (RFI) can significantly pollute synthetic aperture radar (SAR) data and images, which is also harmful to SAR interferometry (InSAR) for retrieving elevational information. To address this issue, in recent years, a class of advanced RFI suppression methods has been proposed based on narrowband properties of RFI and sparsity assumptions of radar echoes or target reflectivity. However, for SAR echoes and the associated scene reflectivity, these assumptions are usually not feasible when the imaged scene is spatially extended. In view of these problems, this study proposes an InSAR-based RFI suppression method for the case of extended scenes. For this task, we combine the RFI-polluted SAR data with RFI-free interferometric data to form an interferometric SAR data pair. We show that such an InSAR data pair embeds an interferogram having the image amplitude multiplying by a complex exponential interferometric phase. We treat the interferogram as a kind of natural image and use discrete Fourier cosine transform (DCT) for its sparse representation. Then combining the DCT-domain sparsity with low-rank modeling of RFI, we retrieve the interferogram and reconstruct the SAR image via joint low-rank and sparse optimization. Numerical simulations show that the proposed method can effectively recover SAR images and interferometric phases from RFI-polluted SAR data.

The problem of strong sea clutter, e.g., sea spikes, may bring in low signal-to-clutter ratio (SCR) and cause great interference to radar marine target detection. However, the sea clutter suppression ability of current algorithms is limited with poor generalization under complex marine environment. In this letter, a novel sea clutter suppression generative adversarial network (SCS-GAN) is designed and employed for marine radar plan-position indicator (PPI) images detection. The SCS-GAN is based on residual networks and attention module, which includes residual attention generator (RAG) and sea clutter discriminator (SCD). In order to expand the data sets and improve generalization ability, clutter-free data set A, simulated sea clutter data set B (containing five types of sea clutter distributions), and actual sea clutter data set C are constructed by means of simulation and acquisition of real radar returns. At last, the parameter, i.e., clutter suppression ratio (CSR) is designed for evaluating the sea clutter suppression performances of the proposed method and other denoising and clutter suppression methods including CBM3D, denoising convolutional neural network (DnCNN), FFDNet, and Pix2pix. After testing with actual data, it is proved that the SCS-GAN has faster clutter removal speed, stronger generalization ability, and at the same time marine targets in images are remained completely.

We propose a drone classification method for polarimetric radar, based on convolutional neural network (CNN) and image processing methods. The proposed method improves drone classification accuracy when the micro-Doppler signature is very weak by the aspect angle. To utilize received polarimetric signal, we propose a novel image structure for three-channel image classification CNN. To reduce the size of data from four different polarization while securing high classification accuracy, an image processing method and structure are introduced. The data set is prepared for a three type of drone, with a polarimetric Ku-band frequency modulated continuous wave (FMCW) radar system. Proposed method is tested and verified in an anechoic chamber environment for fast evaluation. A famous CNN structure, GoogLeNet, is used to evaluate the effect of the proposed radar preprocessing. The result showed that the proposed method improved the accuracy from 89.9% to 99.8%, compared with single polarized micro-Doppler image. We compared the result from the proposed method with conventional polarimetric radar image structure and achieved similar accuracy while having half of full polarimetric data.

Most of the existing ground-penetrating radar-based handheld landmine detectors use algorithms with low computational complexity to obtain detection results quickly. However, these algorithms do not show robust performance against background signals, such as noise signal, shake of sensor, and fluctuation of the signal. To ensure robustness against various background signals, a detection method using a model considering the statistical characteristics of signals should be used. Principal component analysis (PCA), which is a representative method, showed excellent performance in preprocessing and detection steps in the field of landmine detection. However, when the number of training data is small, the performance of PCA cannot be guaranteed. In addition, the amount of computation is increased because of high-dimensional data. Therefore, in this study, we propose a bilateral 2-D-discrete cosine transform PCA (B2D-DCT-PCA) method that shows robust performance with a small number of training samples and reduces computational complexity. In addition, landmine-background reconstruction error (LBRE) is introduced as an index for evaluating how well the difference between landmine and its background is distinguished to determine a suitable preprocessing method for the proposed method, and LBREs according to various preprocessing methods are compared. As a result, the frequency shifting showing the best performance is used. After optimizing the parameters of the proposed method, B2D-DCT-PCA, we confirm that it shows better performance compared with the conventional methods.