Optical image sequences of spacecraft acquired by space-based monocular cameras are typically imaged through exposure bracketing. The feature deformable alignment network for multi-exposure fusion (SFDA-MEF) aims to synthesize a High Dynamic Range (HDR) from set Low (LDR) images with varying exposures. HDR contains details the observed target in LDR captured within specific luminance range. relative attitude camera coordinate system undergoes continuous changes during orbital rendezvous,...

Feature point detection in inverse synthetic aperture radar (ISAR) images of space targets is the foundation for tasks such as analyzing target motion intent and predicting on-orbit status. Traditional feature methods perform poorly when confronted with low texture uneven brightness characteristics ISAR images. Due to nonlinear mapping capabilities, neural networks can effectively learn features from targets, providing new ideas detection. However, scarcity labeled image data presents a...

Accurately reconstructing the intricate structure of natural organisms is long-standing goal 3-dimensional (3D) bioprinting. Projection-based 3D printing boasts highest resolution-to-manufacturing time ratio among all 3D-printing technologies, rendering it a highly promising technique in this field. However, achieving standardized, high-fidelity, and high-resolution composite structures using bioinks with diverse mechanical properties remains marked challenge. The root challenge lies neglect...

Attitude estimation for space targets based on inverse synthetic aperture radar (ISAR) is essential various on-orbit operations, including spacecraft status analysis and fall forecast. However, traditional attitude methods ISAR images suffer from poor feature extraction performance. Furthermore, the present can only deal with stabilized spacecraft. In situations where a noncooperative in slow-spinning state, cross-range scaling inaccurate, resulting incorrect estimation. this study, we...

Instantaneous attitude estimation for spacecraft through optical/ISAR fusion observation is a hot topic in space situational awareness. However, due to the imaging characteristics and domain gap between two data types, it difficult autonomously estimate instantaneous based on fusion. In this paper, an automatic multi-feature framework proposed. We deploy semantic feature extraction networks extract associate kinds of robust features optical ISAR images, algorithm considering cross-range...

Pointing estimation for spacecraft using Inverse Synthetic Aperture Radar (ISAR) images plays a significant role in space situational awareness and surveillance. However, feature extraction cross-range scaling of ISAR create bottlenecks that limit performances current methods. Especially, the emergence staring imaging satellites, characterized by complex kinematic behaviors, presents novel challenge to this task. To address these issues, article proposes pointing method based on...

Cross-range scaling of sequential inverse synthetic aperture radar (ISAR) images is a vital problem for the three-dimensional (3D) reconstruction spatial targets. Specifically, since angular velocity targets cannot be accurately estimated due to their non-cooperative nature, obtained cross-range coordinates scattering points often have non-negligible errors. The errors may result in serious distortion reconstructed geometry targets, which makes consequent target recognition and...

In order to solve the problem of in-shore ship extraction from remote sensing image, a novel method for high resolution (HR) optical image is proposed via salience structure feature and GIS information. Firstly, berth ROI located in with aid prior auxiliary Secondly, salient corner features at bow are extracted precisely. Finally, recursive algorithm concerning symmetric geometry target conducted discriminate multi docked targets into mono ships. The results experiments show that this paper...

This article through the Photoshop to protect environment works production, hope improve everyone's environmental protection consciousness, stimulate students' interest in learning, help teachers teach. The basic operations used work are layer style, highlight, mask, brush tool, cropping invert, torus, filter, channel, decolorization, opacity, etc. Simple and easy understand, meaning of is clear at a glance.This paper first introduces some knowledge software, such as overview, software...

Extracting 3D information from a single image is an interesting but ill-posed problem. Especially for those artificial objects with less texture such as smooth metal devices, the decrease of object detail makes problem more challenging. Aiming at texture-less symmetric structure, this paper proposes novel method pose estimation by introducing implicit structural symmetry and context constraint priori-knowledge. Firstly, parameterized representation, decomposed into series sub-objects regular...

Estimating the 3D pose of space object from a single image is an important but challenging work. Most existing methods estimate known objects and assume that detailed geometry specific known. These are not available for unknown without object. In contrast to previous works, this paper devotes image. Our method estimates only also shape paper, hierarchical model proposed represent prior structure information typical objects. On basis, parameters estimated simultaneously Experimental results...

AbstractThe compressive sensing (CS) based 4-D synthetic aperture radar (SAR) imaging method performs well in the case of high signal-to-noise ratios (SNR). However, presence strong noises, performance CS-based degrades and number false targets increases rapidly. In this paper, a novel SAR is proposed on Bayesian (BCS). Assume that target scattering field follows Cauchy distribution, transformed into signal reconstruction via maximum posteriori estimation. addition, Poisson disk sampling...

The accuracy of 3D viewpoint and shape estimation from 2D images has been greatly improved by machine learning, especially deep learning technology such as the convolution neural network (CNN). However, current methods are always valid only for one specific category have exhibited poor performance when generalized to other categories, which means that multiple detectors or networks needed multi-class object image cases. In this paper, we propose a method with strong generalization ability,...

In this paper, we propose a novel tracking algorithm that can work robustly in small object tracking. The is inspired by recent advances Super-Resolution Convolutional Neural Network (SRCNN). contrast to most existing trackers which simply combine super-resolution (SR) and tracking, deep network was proposed better utilize SR features from different layers for rarely investigated previous works. Specifically, train designed SRCNN offline acquire are more robust against low resolution....

In this letter, a novel method is proposed to conduct sea-land segmentation in the help of geographic information for high resolution (HR) optical remote sensing image. Firstly pre-segmentation conducted by means image analysis. Then salient structure mainly formed artificial facility can be used determine region interests (ROI), which contains potential ship targets. After that, accurate made ROI via mask generated and feature. Experiments on Google shows that paper performs better than...

Recently, convolutional regression networks have drawn great attention in the tracking community. Convolutional trackers formulate network as one layer and take advantages of end-to-end learning. However, existing regress input feature to Gaussian-like soft labels, which still assign a large label semantic backgrounds online model fine-tuning. As result, presence background distractors, tend drift toward regions, exhibit similar appearance compared object interest. In this paper, we propose...

This paper presents a novel behavioral-level analog and mixed-signal (AMS) system performance modeling methodology using support vector machines (SVM). The method relies on linearly graded sub-spaces to model complex multi-dimensional spaces. A detailed evaluation of the has been carried out for purpose potential use AMS synthesis. applied non-ideal 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nd</sup> order sigma-delta modulator (SDM)...

Retracted.

<title>Abstract</title> Assistive wearable systems incorporating soft tactile sensors to transmit spatiotemporal touch patterns have become increasingly prevalent, particularly in biomedical robotics. Although these shown promise tasks such as typing and device operation, their ability convert pressure into specific control commands has been limited, resulting less dexterity intuitiveness than finger-operated electronic devices. Here, we report the fabrication of a oral pad (oPad) equipped...

Unmanned aerial vehicles (UAVs) represent a quickly evolving technology, broadening the availability of remote sensing tools to small-scale research groups in variety scientific fields. Light small UAV is main way obtain centimeter-level high resolution and hour-lever response data. Many technical skills are involved widely development platform covering development, data post-processing, image analysis. Image registration key problem among them. In order improve performance images matching...