Image matching is a primary technology for optical and synthetic aperture radar (SAR) image fusion but often shows limited performance due to the highly nonlinear differences between SAR modalities. Recently, deep neural networks (DNNs) have been investigated effectively extract features tasks, where DNNs are trained based on elaborated design of loss functions low value expected obtain better performance. In this letter, we first theoretically demonstrate that when state-of-the-art function...

Change detection (CD) in optical remote sensing images has significantly benefited from the development of deep convolutional neural networks (CNNs) due to their strong capability local modeling bi-temporal images. In addition, recent rise transformer modules leads improvement global feature extraction Note that existing simple cascade CNNs and shows limited CD performance on small changed areas deficiencies multi-scale information therein. To address aforementioned issue, we propose a new...

Convolutional neural networks (CNNs) have been widely applied in the context of ship detection synthetic aperture radar (SAR) images, but performance is still not ideal scenarios with clutter interference. Mining frequency-domain information to suppress sea SAR has attracted wide attention. However, existing methods do process adaptively, which results degradation performance. To overcome this problem, article proposes a novel deep learning network called YOLO-FA. YOLO-FA contains proposed...

Existing superpixel-based detection algorithms for ship targets in synthetic aperture radar (SAR) images are often derived from the local contrast of intensities (i.e., first-order information superpixels) leading to deteriorating performance low signal-to-clutter ratio (SCR) cases due between and clutter. In this article, we propose a new detector improve target SAR via fisher vectors (LCFVs). The LCFV-based exploits multiorder features superpixels based on Gaussian mixture model (GMM)...

Abstract Metasurfaces, known as ultra‐thin and planar structures, are widely used in optical components with their excellent ability to manipulate the wavefront of light. The key function metasurfaces is spatial phase modulation, originated from meta‐atoms. Thus, find relation between modulation parameters an individual meta‐atom, including sizes, shapes, material's properties, most important but also time‐consuming part metasurface design. Here by developing a backpropagation neural network...

In this letter, we propose a new constant false alarm rate (CFAR) detector to accelerate the existing superpixel (SP)-based CFAR detectors for ship detection in synthetic aperture radar (SAR) images. our method, design density-censoring operation rapidly identify background clutter SPs (BCSPs) with high densities before local detection. way, large number of non-informative BCSPs are removed without time-consuming calculation decision thresholds, and only few candidate target (STSPs)...

This paper proposes a novel unknown input observer (UIO) design method, which incorporates the set-theoretic notions into of UIOs. In this way, we can take advantage both UIOs and methods in fault detection (FD). The main is that they be insensitive to inputs affecting system. However, critical limitation satisfaction UIO conditions for monitored core idea that, even though cannot all inputs, at least as many possible. case, although effect on FD completely removed, partially remove inputs....

In this paper, we propose a two-level block matching pursuit (TLBMP) algorithm based on probabilistic graph model for polarimetric through-wall radar imaging (TWRI). typical L-band to X-band TWRI, indoor targets assume spatial extent and occupy clustered pixels. When sensing is used obtain independent observations, images of can be enhanced within the joint sparsity framework. Toward objective, TLBMP devised exploit both property pattern multiple images. Simulations experimental results data...

We consider the problem of detection sparse stochastic signals with a distributed sensor network. Multiple sensors in network are assumed to observe signals, which share joint sparsity pattern. The Bernoulli-Gaussian (BG) distribution sparsity-enforcing capability is imposed on signals. degree BG model positive and close zero presence absence Motivated by this, formulated as one-sided hypothesis testing degree. For this problem, we propose detector based locally most powerful test (LMPT)...

In this paper, we consider the problem of detection sparse stochastic signals with quantized measurements in sensor networks. The observed are assumed to follow Bernoulli-Gaussian distribution. Due limited bandwidth networks, local sensors required send fusion center. First, propose a detector using locally most powerful test (LMPT) strategy, called LMPT detector, for distributed measurements. Then, quantizers designed guarantee near optimal performance proposed detector. When quantization...

Safe reinforcement learning aims to learn the optimal policy while satisfying safety constraints, which is essential in real-world applications. However, current algorithms still struggle for efficient updates with hard constraint satisfaction. In this paper, we propose Penalized Proximal Policy Optimization (P3O), solves cumbersome constrained iteration via a single minimization of an equivalent unconstrained problem. Specifically, P3O utilizes simple yet effective penalty approach...

The simple linear iterative clustering (SLIC) has been shown as an efficient and widely used superpixel-based algorithm for segmenting marine synthetic aperture radar (SAR) images. However, SLIC does not consider the fact that density of ship target pixels is significantly lower than sea clutter pixels, leading to a waste computational cost memory resources on lots pure areas degradation compactness superpixels. To address aforementioned issues, we develop new density-based (DSLIC) method...

Insufficient reactive oxygen species originating from hypoxia and high glutathione (GSH) in the tumor microenvironment (TME) is an important reason for radiotherapy (RT) resistance. Currently, radiosensitizers that remodel TME are widely investigated to enhance RT. However, developing effective nano‐radiosensitization system removes radiotherapy‐resistant factors boost RT effect while visualizing imaging aid therapy remains a challenge. Herein, MnO 2 nanosheets grown on surface of ultrasmall...

Flood monitoring is of crucial importance for protecting lives and properties. Change detection (CD) methods on multi-source remote sensing images have been widely used flood extent monitoring. In this paper, we propose a spatiotemporal fusion CD (STFCD) algorithm, exploiting the spatial dependence temporal interaction heterogeneous (MSH) satellite image time series (SITS), to realize improved performance in comparison with existing methods. The proposed STFCD algorithm mainly contains two...

With the soaring development of deep learning (DL) mechanisms in recent years, convolution neural network (CNN)-based methods have been extensively investigated to achieve high accuracy ship detection Synthetic Aperture Radar (SAR) images. However, existing CNN-based SAR still suffer from challenges complex inshore scenarios due strong interference therein. To tackle this issue, a novel Semi-Soft Label-guided based on Self-Distillation (SD) for (S <sup...

For the task of change detection (CD) in remote sensing images, deep convolution neural networks (CNNs)-based methods have recently aggregated transformer modules to improve capability global feature extraction. However, they suffer degraded CD performance on small changed areas due simple single-scale integration CNNs and modules. To address this issue, we propose a hybrid network based multi-scale CNN-transformer structure, termed MCTNet, where local information is exploited enhance...

In this paper, we consider distributed detection of sparse stochastic signals with quantized measurements. Assume that both the noise and dominant elements in follow generalized Gaussian (GG) distribution. Since communication bandwidth sensor networks is limited, local sensors send measurements instead analog data to fusion center. First, propose locally most powerful test (LMPT) detector based on for GG Then, quantizers are designed by maximizing efficacy proposed LMPT detector. particular,...

As a powerful coding strategy for superpixels in synthetic aperture radar (SAR) images, Fisher vector (FV) lies low-dimensional subspace and can be sparsely represented as linear combination of training samples. The existing ship detection methods based on FVs often consider the Euclidean distances between target clutter FVs, where features are generally not exploited. In this article, we propose new algorithm nonnegative sparse locality-representation (NSLR) to exploit FVs. proposed NSLR...

Fisher vectors (FVs) can capture multiple order information from superpixels (SPs) in synthetic aperture radar (SAR) images. Existing FV-based ship detectors mainly exploit the local contrast of FVs (LCFVs) but do not consider their global density features. This may lead to degraded performance terms discrimination between targets and complex sea clutter. In this article, two new cues are designed based on fact that target exhibit much lower densities than those clutter also have large...

Conventional compressive sensing (CS) aims at sparse signal recovery from the measurements with continuous values. Quantized CS (QCS) methods arise in digital implementations where quantization of receiver data is performed prior to processing. The extreme case QCS so-called 1-bit each real-valued measurement maintains only sign information one bit. alleviates burden storage and transmission large volumes reduces cost analog-to-digital converter. Recently, has been successfully applied...

We consider the distributed detection of weak signals from one-bit measurements collected by a sensor network where observation model uncertainties exist at all nodes. To solve this problem, locally most powerful test (LMPT) detector is proposed in letter. Moreover, asymptotically optimal quantizers nodes are designed for LMPT detector. In letter, interpreted as multiplicative noise and its variance represents strength uncertainties. Theoretical analysis indicates that, when finite, using...