As the resolution of Earth observation imagery advances, many downstream industries that rely on remotely sensed can also advance. Due to physical limitations optical sensors carried by satellites, growth remote sensing images becomes difficult. Therefore, it is becoming increasingly important boost methods other than upgrading components, such as super-resolution. a computer vision task, convolutional neural networks (CNNs) perform well super-resolution tasks. Transformer-based models show...

To solve the problem of measuring azimuth attitude angle a laser tracking cooperative target at long distance, high-precision measurement method based on beam direction is proposed in this paper. First, optical properties cube-corner prism and principle are described detail. Second, paper proposes for using with an incision, scanning mirror, photodetector. It then analyzes mathematical relationship between vector, photodetector, target, establishes solution model design. Finally, system...

Small UAV target detection and tracking based on cross-modality image fusion have gained widespread attention. Due to the limited feature information available from small UAVs in images, where they occupy a minimal number of pixels, precision required for algorithms is particularly high complex backgrounds. Image techniques can enrich detailed UAVs, showing significant advantages under extreme lighting conditions. registration fundamental step preceding fusion. It essential achieve accurate...

We propose a real-time 3D model-based method that continuously recognizes dimensional emotions from facial expressions in natural communications. In our method, models are restored 2D images, which provide crucial clues for the enhancement of robustness to overcome large changes including out-of-plane head rotations, fast motions and partial occlusions. To accurately recognize emotion, novel random forest-based algorithm simultaneously integrates two regressions tracking continuous emotion...

Millimeter wave (MMW) radar simultaneous localization and mapping (SLAM) technology is an emerging in a tunnel vehicle accident rescue scene. It powerful tool for statistic-trapped detection with limited vision caused by darkness, heat, smoke. A variety of SLAM frameworks have been proven to be able obtain 3-degree-of-freedom (3-dof) pose estimation results using 2-dimention (2D) MMW open space. In the application millimeter closed environment, space structures artificial targets together...

We present a continuous-wave Doppler radar with multi-element receiver array for tracking humans. Joint processing and spatial beamforming are used to resolve multiple movers in the direction of arrival (DOA) space. The improvement performance compared previously developed two-element system is evaluated through Monte Carlo simulation. In an limited size, sidelobes strong targets prevent detection weaker targets. To overcome this limitation, monopulse, CLEAN RELAX algorithms investigated...

In the field of autonomous driving, millimeter-wave (MMW) radar is often used as a supplement sensor other types sensors, such optics, in severe weather conditions to provide target-detection services for driving. RODNet (A Real-Time Radar Object-Detection Network) one most widely MMW range–azimuth (RA) image sequence algorithms based on Convolutional Neural Networks (CNNs). However, adopts an object-location similarity (OLS) detection method that independent number targets obtain final...

Ground-based synthetic aperture radar (GBSAR) has the advantage of retrieving submillimeter deformation mine slope by using differential interferometry technique, which is important for safe production in mining applications. However, moving vehicle’s defocus/displaced signal will mask SAR image area affects accuracy interference phase extraction and inversion. In order to remove its influence, target can first be refocused then removed. To our knowledge, there no GBSAR refocusing method...

Ground-Based Synthetic Aperture Radar (GBSAR) has non-contact, all-weather, high resolution imaging and microdeformation sensing capabilities, which offers advantages in applications such as building structure monitoring mine slope deformation retrieval. The Circular Scanning (CS-GBSAR) is one of its newest developed working mode, the radar rotates around an axis a vertical plane. Such nonlinear observation geometry brings unique advantage three-dimensional (3D) compared with traditional...

For modern radar systems, small unmanned aerial vehicles (UAVs) belong to a typical types of targets with ‘low, slow, and small’ characteristics. In complex combat environments, the functional requirements systems are not only limited achieving stable detection tracking performance but also effectively complete recognition UAV targets. this paper, multi-dimensional feature fusion framework for target utilizing small-sized low-cost high-resolution is proposed, which can fully extract combine...

Radar is a non-contact, high-precision vibration measurement device and an important tool for bridge monitoring. Vibration information needs to be extracted from the radar phase, but phase sensitive noise. Under low signal-to-noise ratio (SNR) data acquisition conditions, such as transmission power or long observation distance, differential jump errors occur clutter estimation becomes difficult, which leads inaccurate inversion of deformation. Traditional low-pass filtering methods can...

The range-instantaneous Doppler (RID) algorithm has been successfully used in high-resolution inverse synthetic aperture radar (ISAR) imaging of maneuvering ship targets. Parametric technique RID is an important method to produce the well-focused ISAR images by estimating parameters received signal modeled as multicomponent polynomial-phase signals each range bin. However, cross terms and false maxima time-frequency distributions cause image blurring loss fine detail. In this paper, we first...

It is necessary to study the influence of atmospheric disturbance on aircraft motion for compensation sensitivity requirements airborne SAR. Most SAR focusing effects assuming errors images from early years development. In this paper, based power spectra density (PSD) method, that containing all response information due can be calculated given turbulence and transfer function between turbulence. As an example, one common model applied get lateral/longitudinal uncontrolled controlled motion....

The single channel Circular Synthetic Aperture Radar (CSAR) has the advantage of continuous surveillance a fixed scene interest, which can provide high frame rate image sequences to detect ground moving targets. Recent image-sequence-based CSAR target detection methods utilize fact that signal moves fast in sequence. Knowledge target’s trace(moving trace sequence, is equal signature’s morphology full aperture image) help design better methods. However, previous signature studies are based on...

Few-shot learning (FSL) is one of the significant and hard problems in field image classification. However, contrast to rapid development visible light dataset, progress SAR target classification much slower. The lack unified benchmark a key reason for this phenomenon, which may be severely overlooked by current literature. researchers always report their new results on own datasets experimental setup. It leads inefficiency result comparison impedes further area. Motivated observation, we...

Millimeter wave radar has the advantage of all-day and all-weather capability for detection, speed measurement. It plays an important role in urban traffic flow monitoring safety monitoring. The conventional 2-dimensional Fast Fourier Transform (2DFFT) algorithm is performed target detection range-Doppler domain. However, motion will induce range walk phenomenon, which leads to a decrease energy performance To solve above problems, this paper proposes moving vehicle based on Keystone-2DFFT...

This paper analyses ambiguity suppression caused by multiple-input multiple-output (MIMO) SAR azimuth nonuniform samplings. Two methods are analyzed: spectrum reconstruction algorithm and minimum mean square error (MMSE) imaging algorithm. The can reconstruct the scene fine resolution, while nonideal orthogonality of multi-channel encoding waveforms causes ambiguous in imaging. MMSE perfectly reconstruct, it requires high SNR.

摘要: 该文在研究星载SAR使用距离向数字波束形成(DBF)实现宽测绘带的基础上，通过理论推导和仿真分析了仅做时域加权DBF处理对成像结果幅度和分辨率的影响，并进一步针对DBF-SAR扫描接收方式研究了4种适用于不同发射信号脉宽的距离向DBF处理方法。仿真验证了4种处理方法在各自的前提下能有效实现距离向宽测绘带接收。 关键词: 合成孔径雷达 / 高分辨率宽测绘带 距离向数字波束形成 扫描接收

Graph Neural Networks (GNNs) have achieved remarkable successes in various graph-based learning tasks, thanks to their ability leverage advanced GPUs. However, GNNs currently face challenges arising from the concurrent use of Tensor Cores (TCs) and CUDA (CDs) These are further exacerbated due repeated, inefficient, redundant aggregations GNN that result high sparsity irregular non-zero distribution real-world graphs. We propose RT-GNN, a framework based on fusion TC CD units, eliminate...