HWHongyuan Wang

Hongyuan Wang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Hongyuan Wang has a total of 59 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    PNG-Net: Physics Noise-Guided Network for Infrared Image Nonuniformity Correction

    by Mei Ming, Haifeng Zhang, Hongyuan Wang

    Infrared thermal imaging is widely used for industrial fault diagnosis because it enables non-contact measurements and reveals temperature anomalies. However, detector nonuniformity degrades image quality and can affect subsequent analysis. Existing nonuniformity correction (NUC) methods face a trade-off between suppressing structured noise and preserving fine image details. To address this issue, a Physics Noise-Guided Network (PNG-Net) is proposed for infrared image nonuniformity correction. First, the input image is decomposed into low-, mid-, and high-frequency components using the fast Fourier transform (FFT), and multi-frequency information is adaptively integrated through a dynamic frequency-band selection module. Second, a learnable PSD-like noise estimator derives a spectral noise prior from the Fourier magnitude spectrum, and an anisotropic mask adds directional information about stripe noise. During feature restoration, the estimated prior noise is incorporated into the proposed Noise-Guided Stack to conditionally modulate multi-scale residual features, thereby progressively recovering structural details and thermal gradients through hierarchical encoder–decoder skip connections. Finally, global residual learning is adopted to predict the nonuniformity correction residual. On the synthetic infrared nonuniformity dataset, PNG-Net achieves a PSNR of 48.92 dB and an SSIM of 0.990, outperforming the compared methods in these two metrics.

  • Open Access

    Dynamic Imaging Simulation and Angular Measurement Performance Degradation of Interferometric Star Trackers

    by Shuai Yao, Hongyuan Wang, Weifeng Du, Zhiqiang Yan, Xunjiang Zheng

    Interferometric star trackers (ISTs) achieve high-precision angular measurement by encoding the stellar incident direction into multichannel interference phases. Besides image smearing, platform motion continuously changes the interference phase during a finite exposure, an effect not captured by conventional geometric star-image models. However, the quantitative relationship between platform motion characteristics and interferometric measurement degradation, which is essential for dynamic performance assessment and system-level optimization of ISTs, remains insufficiently characterized. We develop a full-link dynamic imaging model that incorporates broadband stellar radiation, multiple synthetic diffraction orders, multichannel energy modulation, exposure integration, and detector noise. The four-channel response is expressed as the temporal mean of a complex interference phasor, whose magnitude and argument define the modulation retention factor and dynamic phase bias, respectively. Static angular scanning experiments yield correlation coefficients above 0.97 between simulated and measured channel responses. Under a locally linear phase-to-angle relationship along the interferometric sensing direction, constant-rate motion produces a sinc response, whereas integer-cycle periodic jitter produces a zeroth-order Bessel response. For noninteger-cycle jitter, the response additionally depends on the exposure-to-jitter period ratio and initial phase. When constant-rate motion and periodic jitter coexist, they modulate the same phasor and produce a generally nonseparable response. Within the investigated parameter range, the maximum absolute difference in modulation retention between the coupled response and the independent sinc–Bessel product reaches approximately 0.58. This peak occurs near a normalized phase sweep of 1 and a normalized jitter-induced phase amplitude of 0.8. The model provides a basis for defining dynamic operating limits and selecting exposure parameters and platform stability requirements for ISTs.

  • Open Access

    Performance Evaluation, Optical Optimization and Earth-Based Validation of Star Sensors for Ground Detection in Martian Dust Environments

    by Yuan Gao, Ming-Jian He, Yan Li, Hong-Yuan Wang, Shun-Li Li, Hong Qi

    In deep-space exploration and remote sensing, characterizing radiative transfer in complex planetary atmospheres is fundamental for robust target detection and optical navigation. On the Martian surface, intense scattering and attenuation by dust aerosols pose severe environmental interference, challenging star sensors used for high-precision navigation. To address this, this study develops a spectral radiative transfer model based on the Null Collision Monte Carlo Method to characterize the optical background of the dusty Martian atmosphere. Mie scattering theory is employed for dust particles, while gas molecular absorption is modeled via line-by-line integration. The simulated sky radiance is validated against Mars rover Navcam observations, yielding an average relative error of 7.83% between the modeled and observed radiance values across scattering angles greater than 5°. Building on this, an imaging link model evaluates surface-based detection performance, including signal-to-noise ratio, detection success probability, and star count. Optical parameters—aperture, field of view, and integration time—are optimized for nighttime and dawn-dusk modes. Spatio-temporal assessments are conducted globally across Martian years, focusing on the Zhurong landing site and Tianwen-3 candidates. Finally, an Earth-environment equivalence experiment using a 60% transmittance filter verifies the design’s robustness. This work confirms the feasibility of star-sensor-based attitude determination on Mars.

  • Open Access

    In-Orbit MapAnything: An Enhanced Feed-Forward Metric Framework for 3D Reconstruction of Non-Cooperative Space Targets Under Complex Lighting

    by Yinxi Lu, Hongyuan Wang, Qianhao Ning, Ziyang Liu, Yunzhao Zang, Zhen Liao, Zhiqiang Yan

    Precise 3D reconstruction of non-cooperative space targets is a prerequisite for active debris removal and on-orbit servicing. However, this task is impeded by severe environmental challenges. Specifically, the limited dynamic range of visible light cameras leads to frequent overexposure or underexposure under extreme space lighting. Compounded by sparse textures and strong specular reflections, these factors significantly constrain reconstruction accuracy. While existing general-purpose feed-forward models such as MapAnything offer efficient inference, their geometric recovery capabilities degrade sharply when facing significant domain shifts. To address these issues, this paper proposes an enhanced 3D reconstruction framework tailored for the space environment named In-Orbit MapAnything. First, to mitigate data scarcity, we construct a high-quality space target dataset incorporating extreme illumination characteristics, which provides comprehensive auxiliary modalities including accurate camera poses and dense point clouds. Second, we propose the SatMap-Adapter module to mitigate feature degradation caused by severe specular reflections. This architecture employs a hierarchical cascade sampling strategy to align multi-level backbone features and utilizes a lightweight adaptive fusion module to dynamically integrate shallow photometric cues, intermediate structural information, and deep semantic features. Finally, we employ a weight-decomposed low-rank adaptation strategy to achieve parameter-efficient fine-tuning while strictly freezing the pre-trained backbone. Experimental results demonstrate that the proposed method decreases the absolute relative error and Chamfer distance by 15.23% and 20.02% respectively compared to the baseline MapAnything model, while maintaining a rapid inference speed. The proposed approach effectively suppresses reconstruction noise on metallic surfaces and recovers fine geometric structures, validating the effectiveness of our feature-enhanced framework in extreme space environments.


  • Open Access

    Research on Star Sensor Imaging Simulation Under Near-Space Hypersonic Non-Equilibrium Flow Conditions

    by Zhen Liao, Hongyuan Wang, Xi Cheng, Boqi Liu, Yunzhao Zang, Yinxi Lu, Shuai Yao, Zhiqiang Yan

    In order to address the difficulty in acquiring degraded images for star sensors under hypersonic conditions, this paper proposes a star sensor imaging simulation model which can effectively analyze the influence of thermochemical non-equilibrium flow on the star sensor. Firstly, the two-temperature model is adopted to describe the relaxation phenomenon of hypersonic non-equilibrium flow, and the chemical reaction is simulated by Arrhenius law. Then, the effects of optical transmission and thermal radiation on star sensor imaging are quantitatively analyzed. Based on this degradation model, the degraded star images under two typical working conditions are simulated. The simulation results show that the radiation of the solid media has the most significant influence on the imaging of the star sensor. The detectable limit magnitudes of the star sensor obtained under the two working conditions are 3.28 and 4.55, respectively. The research results can provide important theory and technical support for the system design and algorithm testing of star sensors on near-space hypersonic platforms.

  • A Phase-Compensation-Based Method for Improving the Centroiding Accuracy of Interferometric Star Trackers

    by Shuai Yao, Hongyuan Wang, Xunjiang Zheng, Zhiqiang Yan, Zhen Liao, Yunzhao Zang

    For interferometric star trackers operating under broadband conditions, the zeroth-order diffraction from the Ronchi grating significantly degrades centroiding accuracy. To address this, we present a solution based on a hardware–software codesign. First, we systematically investigate this issue and establish a stellar interference imaging model using the angular spectrum propagation theory. The model accounts for stellar radiation characteristics, grating diffraction properties, and the performance of the beam-splitter assembly. Guided by this model, we introduce a phase-compensated multiwedge beam-splitter architecture together with a centroiding algorithm specifically tailored to the resulting interference modulation. This joint optical algorithm design effectively suppresses the zeroth-order diffraction component at the physical level and maximizes the utilization of the compensated interference signal. Numerical simulations incorporating detector noise are conducted to obtain the star spot energy distribution for centroid evaluation. Simulation results show that, along the grating interference direction, the mean absolute error (MAE) of centroid localization decreases from 0.169 to 0.147 arcsec, while the root-mean-squared error (RMSE) decreases from 0.211 to 0.186 arcsec, indicating a significant improvement in overall accuracy. The proposed method effectively mitigates the adverse effects of the zeroth-order diffraction and serves as an important reference for the design and performance optimization of interferometric star trackers.


  • Fully Covered Adversarial Camouflage Against Remote Sensing Detection via Physics-Driven Rendering and Pyramid Training

    by Ziyang Liu, Hongyuan Wang, Qianhao Ning, Zijian Wang, Yinxi Lu, Yunzhao Zang, Zhiqiang Yan

    Unmanned aerial vehicle (UAV) imaging object detection systems based on deep neural networks (DNNs) are highly vulnerable to adversarial attacks. However, existing methods have not fully considered the complex physical interferences in the real world, nor the multi-scale and temporal observation characteristics of UAV-based detection, which makes the generated camouflage difficult to maintain effectiveness in practical tasks. To address this limitations, a fully-covered robust adversarial camouflage generation framework for UAV-based detection models is introduced. First, the Physically Based Differentiable Renderer (PBDR) is designed to simulate global illumination, surface reflectance, and atmospheric scattering, thereby incorporating complex physical interference factors during camouflage training. Second, the Pyramid Camouflage Training Framework (PCTF) is introduced, which conducts hierarchical training on distance-varying inputs to preserve the multi-scale adaptability of camouflage features across layers and ensure inter-frame consistency within layers. Furthermore, the limitations of existing evaluation indicators for UAV-based adversarial attacks are analyzed, and a novel set of indicators is proposed for evaluating physical adversarial camouflage in remote sensing. Extensive cross-domain experiments in both digital space and physical world demonstrate superior performance compared with SOTA methods. The proposed approach achieves a temporal attack success rate (TASR) of 84.5%1, compared with baselines, the average precision (AP) scores across multiple black-box detection models decrease by 18.4% on average, while the attack success rate (ASR) increases by 70.4%.


  • Naturalistic physical adversarial camouflage for object detection via differentiable rendering and style learning

    by Ziyang Liu, Zhiqiang Yan, Qianhao Ning, Yinxi Lu, Zijian Wang, Hongyuan Wang


  • Open Access

    Noncooperative Spacecraft Pose Estimation Based on Point Cloud and Optical Image Feature Collaboration

    by Qianhao Ning, Hongyuan Wang, Zhiqiang Yan, Zijian Wang, Yinxi Lu

    Pose estimation plays a crucial role in on-orbit servicing technologies. Currently, point cloud registration-based pose estimation methods for noncooperative spacecraft still face the issue of misalignment due to similar point cloud structural features. This paper proposes a pose estimation approach for noncooperative spacecraft based on the point cloud and optical image feature collaboration, inspired by methods such as Oriented FAST and Rotated BRIEF (ORB) and Robust Point Matching (RPM). The method integrates ORB feature descriptors with point cloud feature descriptors, aiming to reduce point cloud mismatches under the guidance of a transformer mechanism, thereby improving pose estimation accuracy. We conducted simulation experiments using the constructed dataset. Comparison with existing methods shows that the proposed approach improves pose estimation accuracy, achieving a rotation error of 0.84° and a translation error of 0.022 m on the validation set. Robustness analysis reveals the method’s stability boundaries within a 30-frame interval. Ablation studies validate the effectiveness of both ORB features and the transformer layer. Finally, we established a ground test platform, and the experimental data results validated the proposed method’s practical value.


  • Open Access

    A Fast and Robust Range Alignment Method for ISAR Imaging Based on a Deep Learning Network and Regional Multi-Scale Minimum Entropy Method

    by Qianhao Ning, Hongyuan Wang, Zhiqiang Yan, Zijian Wang, Yinxi Lu

    In inverse synthetic aperture radar (ISAR) imaging, range alignment (RA) is crucial for translational compensation. To address the need for rapidity and accuracy in the RA process, a fast and robust range alignment method is proposed based on a deep learning network and minimum entropy (ME) method. The proposed method consists primarily of two components: the CNN-RNN attention mechanism network (CRAN) architecture and the regional multi-scale minimum entropy (RMSME) method. The main distinction of this method from existing approaches lies in its utilization of a deep learning network for rapid coarse alignment, followed by the search for minimum entropy within local regions at multiple scales. The integration strategy effectively addresses the current challenges of poor generalization in deep learning networks and low efficiency in the traditional ME method. The experimental results of simulation data indicate that the proposed method achieves the best range alignment performance compared to RNN, CRAN, and the traditional ME method. The experimental results of the measured data further validate the practicality of the proposed method. This research provides reference significance for the joint application of deep learning and traditional methods in the RA process.