Cross-domain few-shot scene classification (CDF-SSC) aims to establish cross-domain representation between source and target domains, endowing the model few-shot classification ability on the target domain. Recent studies have improved cross-domain representation learning by incorporating unlabeled target data into semi-supervised training with labeled source data. However, these methods struggle to effectively bridge the domain gap between source and target domains, and lack the discriminative feature description ability for unlabeled target domain data, leading to inferior cross-domain representation learning and affecting the few-shot performance. In this article, a dual-domain representation modeling with prototype contrastive learning (DMPC) structure is proposed to improve the robustness of cross-domain representation learning. In DMPC, first, a dual-domain Gaussian representation modeling is designed to model the feature statistics of both source and target data as multivariate Gaussian distributions rather than fixed values and enrich domain representation by random sampling new feature statistics. It helps bridge the domain gap at the feature level, and improves the model’s robustness and generalization to better address unpredictable variations in the target domain. Second, a pseudo-prototype contrastive learning branch is proposed to improve the discriminability of representation for limited unlabeled target data. By leveraging pseudo-prototypes derived from the classifier’s weights as dynamic anchors, it refines feature representation by clustering features of the same pseudo-class and separating those of different pseudo-classes, strengthening the model’s ability to capture distinct and consistent features within the target domain. Finally, the classifier is fine-tuned on few-shot tasks to adapt to specific categories of the target domain. Extensive experimental results exhibit impressive performance of DMPC on 12 RS cross-domain scenarios.

A unified object detector needs to integrate various abilities for adapting to different remote sensing object detection tasks. However, there is a lack of a feasible way to integrate multigrained object detection requirements i.e., horizontal bounding box (HBB), oriented bounding box (OBB), and instance segmentation (InSeg) into a unified detection way. Then, it often has to design specific parametric learning ways and their corresponding architectures, which cannot be finely adaptive to various kinds of object detection tasks. Therefore, in this article, a new benchmark is set up to integrate multigrained object detection requirements of HBB, OBB, and InSeg into one challenging task of arbitrary-shaped object contour detection. At the same time, a unified object contour detector (UniconDet) is proposed for achieving multigrained object detection from complicated remote sensing scenes. First, a Fourier contour parametric modeling (FCPM) is defined to project arbitrary-shaped object contours from the spatial domain into the frequency domain. Then, it can unify spatial parametric representations of HBB, OBB, and InSeg as frequency coefficient representations, which can be used for realizing a more generic and robust parametric regression. Second, a multiview cross-attention (MVCA) feature extraction way is designed at each scale of the regression layer, which can assist UniconDet in perceiving Fourier contour parameters by exploring the coupled relations between different discrete contour sampling periods of each object. Third, a center-contour enhancing regression layer (C2-ERL) is designed to generate regional guidance and cascade contour propagation, which can ensure a more accurate center point prediction and Fourier contour parameter regression. Finally, extensive experiments are carried out on benchmarks of HBB, OBB, InSeg, and new multigrained object detection, and the results indicate that our proposed UniconDet can obtain superior performance. The source code is available athttps://github.com/ZhAnGToNG1/UniconDet.

Few-shot object detection (FSOD) has to learn classification and localization information for unseen object detection under very low-data resource regimes. However, when deficient samples are adopted for model training, it is hard to build powerful location-aware and identification abilities for well coping with agnostic bias from diverse testing scenarios; at the same time, the overfitting phenomenon is easily occurring. Therefore, in this article, a controllable generative knowledge-driven FSOD called CGK-FSOD is proposed for unseen object detection from optical remote sensing imagery. Specifically, to enrich the learnable data space of scarce samples for preventing incomplete agnostic-bias learning, while avoiding the overfitting phenomenon, a visual-textual prompt-based controllable data generation is designed to generate high-quality object detection data based on pretrained foundational models [i.e., the stable diffusion (SD) and contrastive language-image pre-training (CLIP)], which not only can introduce the generalized domain-level knowledge into the remote sensing domain but also sets up an all-round data space to support complete learning of potential agnostic bias. Furthermore, with respect to the denoising generative process of SD, a series of cross-modality generative features in latent representation space are reused for few-shot fine-tuning by the designed cross-modality feature embedding (CMFE), which not only can bring diverse generative abilities into the feature fusion step of the detector but also gracefully sets up feature representation scalability to make the detector better adapt to agnostic bias from diverse testing scenarios of FSOD. Finally, extensive experiments are executed on two public remote sensing datasets (e.g., DIOR and NWPUVHR-10), and the results indicate that the proposed CGK-FSOD is very effective and flexible for FSOD.

We propose DiffusionEMIS, a new paradigm that formulates the 3-D electromagnetic inverse scattering (EMIS) as a denoising diffusion process from a 4-D noise distribution to the specific 3-D scatterer point cloud with reasonable electrical parameters at each point under the scattering field as a condition to enrich the research of 3-D EMIS community while avoiding the computational burden of 3-D meshes. During the training process, scatterer point clouds diffuse from ground-truth shapes to random distributions, and the model learns to reverse this noising process. During the inference process, the model progressively refines the randomly generated shape to the output result. We design a certain 3-D scattering system to construct two datasets: 3DEMIS-Mixed National Institute of Standards and Technology datab (MNIST), where the samples are nonuniform 3-D handwritten digits, and 3DEMIS-ShapeNet: An information-rich 3-D model repository (SHAPENET), where the samples are uniform 3-D objects. The extensive evaluations on 3DEMIS-MNIST and 3DEMIS-SHAPENET show that our DiffusionEMIS achieves favorable performance and great stability which means 3-D EMIS can be solved by a generative way. Our model outperforms previous common EMIS methods, such as the born iterative method (BIM) by a large margin and has extremely robust noise resistance. To evaluate the generalization performance of DiffusionEMIS, we construct the dataset 3DEMIS-English Mixed National Institute of Standards and Technology database (EMNIST) consisting of nonuniform 3-D handwritten letters for testing the model trained on 3DEMIS-MNIST, the test results indicate a great consistency with the ground truths. We also conducted tests on real data to further demonstrate the effectiveness of our method.

Few-shot remote sensing scene classification (FSRSSC) aims to identify unseen classes only relying on very limited training samples. However, scarce training samples are insufficient to support a robust classwise representation, which is easily influenced by agnostic biases from diverse testing scenarios. Fortunately, there are abundant spatial contextual clues that exist in very limited training samples to have an enormous potential to establish discriminative and transferable concepts. Thus, in this article, a hybrid architecture called ProtoConViT is proposed to learn a powerful classwise representation based on spatial contextual clues for FSRSSC promotion. First, support-query correlative guidance is designed to generate more stable spatial connections among support and query data based on intermediate convolution neural network (CNN) feature maps, which not only can be embedded into each episodic training task to reduce redundant spatial contextual representation learning space of vision transformer (ViT) but also can assist it in rapidly capturing critical spatial contextual clues to classify query data into one of classes from support set. Second, followed by the designed support-query correlative guidance, a novel heterogeneous prototype distillation is proposed to integrate the advantages of CNN and ViT for heterogeneous prototype construction, which can rapidly set up discriminative and transferable concepts for FSRSSC. Third, corresponding to the proposed ProtoConViT, a joint loss is designed to make the model rapid convergence based on meta-learning. Finally, extensive experiments are carried out on three FSRSSC benchmarks, and comparative results indicate that the proposed ProtoConViT can achieve a superior FSRSSC performance.

Recently, semantic segmentation technology has been a research hotspot in optical remote sensing urban use classification. However, because of coupled semantic relations in very high-resolution and complex urban scenes, a more effective semantic description for pixelwise urban use interpretation has become a challenge. Then, aiming to set up a more effective semantic description, the effective receptive field (ERF) is analyzed in general convolutional neural networks. The unreasonable ERF distribution in the stacked convolutional layers of the encoder would lead to a large amound of small ERFs and fewer not large enough ERFs that form a naive semantic description in decoder. Therefore, in this article, a novel full semantic constructed network (FSCNet) is proposed to improve the naive semantic description and set up an effective semantic description. First, to avoid noise from shallow feature layers, a residual refinement convolution is designed to optimize the full-scale skip connections based on the U-shaped encoder–decoder. Second, an interscale fusion module is newly designed for multiscale feature fusion, which can generate three initial semantic modalities that are prepared for redefining the full semantic description. Third, a multiscale local context spatial attention module and boundary supervision are designed for an initial shallow semantic modality to capture the pure boundary information, and then, pyramid spatial pooling is employed for an initial deep semantic modality to further enlarge the ERF and obtain more abstract global information. Next, a self-calibration convolution combined with the atrous spatial pyramid pooling is designed to rectify and enrich an initial middle semantic modality, which can improve the naive semantic description and bridge the semantic gap between the redefined shallow and deep semantic modalities to advance the full semantic feature fusion. Finally, extensive experiments are carried out on three benchmarks (e.g., ISPRS Vaihingen, Potsdam, and DLRSD), and comparative results show that the proposed FSCNet can get remarkable performance compared to state-of-the-art (SOTA) methods. Besides, the code is available athttps://github.com/DorisCV/FSCNet.

It is of great importance to remotely sense trapped victims with radio signals in modern search and rescue after natural disasters like earthquakes, avalanches, building collapses, and so on. Various radio sensors have been developed to date; however, they are hardly deployed to recognize efficiently the vital sign in long-distance, deep-coverage, and multi-subject situations because the back-scattered victim-critical radio signals are really weak and are nearly drowned in the ambient nonstationary noise and clutter. To tackle the formidable difficulty, we present a four-dimensional wideband microwave radar operating at 1.7 GHz to 2.7 GHz and develop a spatio-temporal processing algorithm to fully explore the vital knowledge of victims in the three-dimensional spatial and one-dimensional temporal information. We conducted comprehensive field experiments in real post-disaster environments and demonstrated experimentally that our radio sensor can continuously monitor multiple survivors trapped under mounds of debris in real urban environments. Moreover, we demonstrate that the presented method can achieve a signal-to-noise-and-clutter ratio improvement of more than 20 dB even in the case of deep burial, which enables localizing the victims trapped in the order of ten meters and recognizing the survivors’ vital states. We expect that the presented strategy may open an avenue for future remote life-rescuing and beyond in practical applications.

Arbitrary-oriented object detection (AOOD) from optical remote sensing imagery has to correctly generate delicate oriented boundary boxes (OBBs) and meanwhile identify their specific categories. However, how to make detectors learn delicate parameters of OBBs, especially for the crucial orientation information, and identify object category from complex background becomes a challenge task. Therefore, in this article, for exploring a better way to guide the detector to learn specific category and parametric information of OBBs, a novel one-stage anchor-free detector called Posterior Instance Injection Detector (PIIDet) is proposed for AOOD. First, as the anchor-free manner lacks prior information, an object-aware posterior guidance (OAPG) structure is proposed to generate specific-category instances used for conditioning on OBB prediction. This structure can assist the proposed PIIDet in better learning the relative parametric information of OBBs corresponding to their specific categories. Besides, to guarantee a high quality injection of specific-category instances, a new hierarchical feature fusion module is developed to establish a suitable multi-scale feature mapping space. Second, considering the negative optimization of angle regression, which is caused by the boundary discontinuity of angular periods and sudden shifts of the relation between width and height in training phase, a novel binary classification embedded angle regression space (BCE-RegSpace) is devised for providing continuous angle regression space and stable relation between width and height. Finally, extensive experiments are executed on three AOOD benchmarks (e.g., DOTA, DIOR-R and HRSC2016), and results proved that the proposed concise one-stage anchor-free PIIDet can reach the state-of-the-art (SOTA) performance and meanwhile have an impressive inference speed.

Small vehicle detection is one of the remaining challenging task because the ambiguous appearance is against complex background interference. Consequently, in order to improve the performance of small vehicle detection from aerial optical remote sensing images, a novel adaptive local context (ALC) embedding way is designed and further introduced into an anchor free detection manner which is called ALC-Net, and in ALC-Net, it can adaptively set up the effective local context feature to improve keypoint description of small vehicles and boost the detection performance without adding extra prior information. Finally, several experiments are carried out on two widely used datasets (e.g., UCAS-AOD [1] and VEDAI [2]) and the results indicate that the proposed ALC-Net can exhibit the competitive small vehicle detection performance than other detectors.

Object detection is an essential task in computer vision. Recently, several convolution neural network (CNN)-based detectors have achieved a great success in natural scenes. However, for optical remote sensing images with a large scale of view, lower proportion of foreground target pixels and drastic differences in object scale present considerable challenges. To address these problems, we propose a novel one-stage detector called the full-scale object detection network (FSoD-Net) which consists of proposed multiscale enhancement network (MSE-Net) backbone cascaded with scale-invariant regression layers (SIRLs). First, MSE-Net provides the multiscale description enhancement by integrated the Laplace kernel with fewer parallel multiscale convolution layers. Second, SIRLs contain three different isolated regression branch layers (i.e., corresponding to small, medium, and large scales), which make default discrete scale bounding boxes (bboxes) cover full-scale object information in regression procedure. A novel specific scale joint loss is also designed that uses the softmax function combined with a strong$L_{1}$-norm constraint in each regression branch layer. It can further speed up the convergence and improve the classification scores of predicted bboxes. Finally, extensive experiments are carried on challenge data sets of large-scale dataset for object detection in aerial images (DOTA) and object detection in optical remote sensing images (DIOR) which contain multiple instances from different imaging platforms, and these results demonstrate that FSoD-Net can achieve better performance than other state-of-the-art one-stage detectors, and it can reach a mean average precision (mAP) of 75.33% on DOTA and 71.80% mAP on DIOR, respectively. Especially, the average precision (AP) of tiny object detection can improve 10%–20% approximately.

Optical remote sensing object detection is a challenging task, because of the complex background interference, ambiguous appearances of tiny objects, densely arranged circumstances, and multiclass object with vaster scale variances and irregular aspect ratios. The performance of object detection is seriously restricted. Thus, in this article, inspired by the anchor-free object detection framework, and aiming to solve these difficulties to improve the optical remote sensing object detection performance, a powerful one-stage detector of multiscale semantic fusion-guided fractal convolution network (MSFC-Net) is proposed. First, facing these strong-coupled semantic relations in each complex scene, a compound semantic feature fusion (CSFF) way is designed for generating an effective semantic description, which is a benefit to pixel-wise object center point interpretation. In addition, it can be easily extended into a semantic segmentation task. Second, in view of accurate multiclass pixel-wise center point predictions based on an effective compound semantic description, a novel fractal convolution (FC) regression layer is designed, which adaptively achieves the regression of multiscale bounding boxes (bboxes) with irregular aspect ratio under no priori information. Third, related to the set up FC regression layer, a specific hybrid loss is designed to make the proposed MSFC-Net converge better. Finally, the extensive experiments on challenge data sets of large-scale dataset for object detection in aerial images (DOTA) and object detection in optical remote sensing images (DIOR) datasets are carried out, and comparisons indicate that the proposed MSFC-Net can perform the remarkable performance than other state-of-the-art one-stage detectors, as it can reach 80.26% mean average precision (mAP) and 79.33% mF1 on DOTA and 70.08% mAP and 73.45% mF1 on DIOR. Then, our work is available athttps://github.com/ZhAnGToNG1/MSFC-Net.

Saliency analysis is an essential task in computer vision and aims to generate distinguishing foreground features from background features. However, due to complex structure distributions in large-scale optical remote sensing scenes, generating effective feature descriptions for regional saliency analysis is challenging. Therefore, in this study, we proposed a novel global-local feature assembly method called GLFA-Net based on a convolution neural network (CNN) that can adaptively learn an effective feature representation for regional saliency analysis to achieve the region-of-interest (ROI) (e.g., aircraft carrier, airport, and urban area) extraction from complex optical remote sensing images. In addition, we also collected these complex optical remote sensing scene images from Google Earth and DOTA data sets to demonstrate the effectiveness of the proposed method. Finally, experimentation shows that the proposed regional saliency analysis method can produce better ROI extraction performance than other methods, reaching a 0.057 mean absolute error (MAE), a 0.703 Kappa coefficient, and a 0.735 F1-score.

Using radio-frequency (RF) sensing techniques for human posture recognition has attracted growing interest due to its advantages of pervasiveness, contact-free observation, and privacy protection. Conventional RF sensing techniques are constrained by their radio environments, which limit the number of transmission channels to carry multi-dimensional information about human postures. Instead of passively adapting to the environment, in this paper, we design an RF sensing system for posture recognition based on reconfigurable intelligent surfaces (RISs). The proposed system can actively customize the environments to provide desirable propagation properties and diverse transmission channels. However, achieving high recognition accuracy requires the optimization of RIS configuration, which is a challenging problem. To tackle this challenge, we formulate the optimization problem, decompose it into two subproblems, and propose algorithms to solve them. Based on the developed algorithms, we implement the system and carry out practical experiments. Both simulation and experimental results verify the effectiveness of the designed algorithms and system. Compared to the random configuration and non-configurable environment cases, the designed system can greatly improve the recognition accuracy.

Three-dimensional multistatic imaging is a powerful noninvasive examination tool for many military and civilian applications. Recently, the sparsity-regularized optimization has been used as a popular imaging technique to enhance the image quality. However, it suffers from the expensive computational cost, since its solution is obtained by a time-consuming iterative scheme, which is typically computationally prohibitive for large-scale imaging problems. To overcome this difficulty, this challenging imaging problem is converted into an image processing problem in this letter, which can be performed over small-scale overlapping patches and be efficiently solved in a parallel or distributed manner. In this way, the proposed qualitative scheme could be utilized to solve large-scale imaging problems. Exemplary simulation results are provided to demonstrate the efficiency of the proposed methodology.

This letter introduces the concept of regression structural similarity index measure (regression SSIM) and builds a supervised graph framework of automatic detection of small-size ultrawideband (UWB) radar targets. The twofold contribution made in this letter includes the following: 1) The regression SSIM is proposed to measure the similarity of the local pattern between a test image and a reference image; and 2) the framework of a supervised graph, together with the regression SSIM, has been developed to address the automatic detection of UWB radar objects. As opposed to other detection techniques reported in the literature, our methodology does not rely on statistical modeling or imposing typical shape parameters. Selected results of processing simulated and real UWB ground-penetrating radar data are provided, which verifies the state-of-the-art performance of the proposed methodology and provides important potential for a wide class of target detection.

This paper presents a novel approximate solution for electromagnetic scattering by planar homogeneous dielectric objects with arbitrary shape. This solution is derived by introducing a semi-empirical formula of the electrical field inside objects under consideration and by exploring the volumetric integral equation along with spectral representation of dyadic Green's function. Afterward, a closed-form approximate solution has also been established by utilizing the so-called stationary phase method (SPM). It is demonstrated that our solution is uniformly valid for planar objects with moderate electrical thickness illuminated from low to high frequencies at all incidence angles. Comparisons with a full-wave computation obtained by the method of moments (MoM) have been examined, which demonstrates that the proposed formulation can provide very accurate results and can perform much better than other existing approximate techniques.

Over the past decades, numerous efforts have been attempted to enhance the temporal resolution and accuracy of subsurface ground-penetrating radar (GPR) data by means of blind deconvolution techniques. The fact that most of the source wavelets that are utilized in practical GPR exploration are nonminimum phase presents some challenges for the blind deconvolution of GPR data. This letter extends the classical minimum entropy deconvolution strategy and forms a general-purpose framework of the blind deconvolution of GPR data, which formulates the blind deconvolution of GPR data as a sparsity-promoted optimization problem with a scale-invariant regularizer. Another contribution of this letter is that an alternating iterative method is explored to solve the derived nonconvex optimization problem, where the constraint of maxt|r(t)| = 1 is introduced to avoid trapping into some local minimums. Selected examples are presented to demonstrate the accuracy and robustness of the proposed methodology. Primary results show that by applying such approach to the GPR data, we obtain images with significantly enhanced temporal resolution compared with the results of existing blind deconvolution schemes.

The analysis of synthetic aperture radar (SAR) migration developed by Gilmore has been refined within the context of the inverse scattering problem, particularly the distorted-wave Born approximation (DWBA). The SAR migration algorithm can be deduced from the DWBA-based inversion formulation when the following assumptions are satisfied: 1) homogeneous and nonfrequency-dependent background medium; 2) the exploding source model; and 3) the well-resolved targets described by an orthogonal relation derived in this paper. In addition, the other contributions of this paper are as follows: 1) The removal of the ¿2term has been clarified by the derived orthogonal relation; 2) a scale factor that balances the near-far field has been derived; and 3) a novel SAR migration algorithm for the imaging of targets embedded in a layered medium has been proposed.

A novel real-time through-wall imaging (TWI) algorithm with autofocusing ability in the presence of wall ambiguities is proposed in this paper. The spectrum Green's function is employed to formulate the TWI algorithm, where the fast Fourier transform can be used to reconstruct the image in a very short computation time. The complex scattering process due to the presence of the wall is automatically included in the imaging formulation through the multilayer Green's function. The autofocusing is achieved by introducing a time factor in the TWI formulation to get a dynamic image at different focusing time. The image at the time instant when the defined entropy is minimized is stored as the output of the TWI result. Simulation results show that the proposed method can provide high-quality focused image in a short computation time regardless of the estimated value of the wall parameters.

The closed-form solution to the reconstruction of radiating current with a line measurement configuration is presented in this paper. The analytical result of the continuous singular value decomposition of the scattering integral operator is derived and used to analyze the imaging resolution and reconstruct the radiating component of the equivalent current density. The main four advantages of the proposed closed-form solution are as follows: 1) The reconstruction of radiating current can be achieved in a very short computation time; 2) it is quite tolerant to different levels of noises; 3) it is very easy to realize the reconstruction of radiating currents within the obstacles embedded in layered medium; and 4) the imaging resolution kernel of the scattering operator can also be derived in a closed form. Numerical simulations show the high efficiency of the proposed method for the reconstruction of radiating current.

In this paper, two new approaches are presented for the solution of electromagnetic inverse scattering problems when amplitude-only data are available. The proposed techniques are based on a customized version, which are the so-called contrast source inversion (CSI) and multiplicative regularized CSI (MRCSI) methods. In the proposed approaches, denoted as the phaseless-data (PD)-CSI and the PD-MRCSI, only the term of the cost functional concerning the mismatch between the measured and estimated field data (i.e., the data equation) has been properly redefined. Moreover, the back-projection algorithm has been modified to provide an initial solution ensuring the rapid convergence of the optimization procedures and avoid the reconstruction of false solutions. A set of representative results concerning numerical as well as experimental tests is reported to show the accuracy of the proposed amplitude-only reconstruction approaches.

Inspired by the successes of the distorted Born iterative method with full data (FD-DBIM) in the electromagnetic inverse scattering problem when the weak scattering approximation breaks down, the distorted Rytov iterative method with phaseless data (PD-DRIM) for the reconstruction of an unknown object embedded in 2-D homogeneous space is proposed when only the intensity data of the total wavefield are available. Similar to FD-DBIM, the PD-DRIM algorithm can also provide the Fre acutechet derivative and the guess solution for the iterative method naturally by updating the Green's function in every iteration step. The validation of the proposed approach is justified by the experimental tests where the scattering data are kindly provided by the Institute Fresnel in Marseille (2005).

The effects of ionospheric irregularities on spaceborne synthetic aperture radar (SAR) signal propagation with double path and multilook angle are studied in a model in which a two-frequency and two-position coherence function (TFTPCF) has been adopted for analysis. The TFTPCF is derived from the phase-screen principle. The ambiguity function based on TFTPCF has been used to analyze the effects of ionospheric turbulence on range resolution and cross resolution. The results show that, in some cases, the effects from the irregularities on SAR imaging can be very serious