arXiv subjects

Liu Wei

Publications and source records attributed to Liu Wei.

9 recordsLinked to original sources

arXiv2026

UAV Trajectory Optimization via Improved Noisy Deep Q-Network

This paper proposes an Improved Noisy Deep Q-Network (Noisy DQN) to enhance the exploration and stability of Unmanned Aerial Vehicle (UAV) when applying deep reinforcement learning in simulated environments. This method enhances the exploration ability by combining the residual NoisyLinear layer with an adaptive noise scheduling mechanism, while improving training stability through smooth loss and soft target network updates. Experiments show that the proposed model achieves faster convergence and up to $+40$ higher rewards compared to standard DQN and quickly reach to the minimum number of steps required for the task 28 in the 15 * 15 grid navigation environment set up. The results show that our comprehensive improvements to the network structure of NoisyNet, exploration control, and training stability contribute to enhancing the efficiency and reliability of deep Q-learning.

eess.SY↗

arXiv2025

Real-time deep learning phase imaging flow cytometer reveals blood cell aggregate biomarkers for haematology diagnostics

While analysing rare blood cell aggregates remains challenging in automated haematology, they could markedly advance label-free functional diagnostics. Conventional flow cytometers efficiently perform cell counting with leukocyte differentials but fail to identify aggregates with flagged results, requiring manual reviews. Quantitative phase imaging flow cytometry captures detailed aggregate morphologies, but clinical use is hampered by massive data storage and offline processing. Incorporating hidden biomarkers into routine haematology panels would significantly improve diagnostics without flagged results. We present RT-HAD, an end-to-end deep learning-based image and data processing framework for off-axis digital holographic microscopy (DHM), which combines physics-consistent holographic reconstruction and detection, representing each blood cell in a graph to recognize aggregates. RT-HAD processes >30 GB of image data on-the-fly with turnaround time of <1.5 min and error rate of 8.9% in platelet aggregate detection, which matches acceptable laboratory error rates of haematology biomarkers and solves the big data challenge for point-of-care diagnostics.

q-bio.QM↗

arXiv2022

A Novel Image Descriptor with Aggregated Semantic Skeleton Representation for Long-term Visual Place Recognition

In a Simultaneous Localization and Mapping (SLAM) system, a loop-closure can eliminate accumulated errors, which is accomplished by Visual Place Recognition (VPR), a task that retrieves the current scene from a set of pre-stored sequential images through matching specific scene-descriptors. In urban scenes, the appearance variation caused by seasons and illumination has brought great challenges to the robustness of scene descriptors. Semantic segmentation images can not only deliver the shape information of objects but also their categories and spatial relations that will not be affected by the appearance variation of the scene. Innovated by the Vector of Locally Aggregated Descriptor (VLAD), in this paper, we propose a novel image descriptor with aggregated semantic skeleton representation (SSR), dubbed SSR-VLAD, for the VPR under drastic appearance-variation of environments. The SSR-VLAD of one image aggregates the semantic skeleton features of each category and encodes the spatial-temporal distribution information of the image semantic information. We conduct a series of experiments on three public datasets of challenging urban scenes. Compared with four state-of-the-art VPR methods- CoHOG, NetVLAD, LOST-X, and Region-VLAD, VPR by matching SSR-VLAD outperforms those methods and maintains competitive real-time performance at the same time.

cs.CV↗

arXiv2021

Two Chebyshev Spectral Methods for Solving Normal Modes in Atmospheric Acoustics

The normal mode model is important in computational atmospheric acoustics. It is often used to compute the atmospheric acoustic field under a harmonic point source. Its solution consists of a set of discrete modes radiating into the upper atmosphere, usually related to the continuous spectrum. In this article, we present two spectral methods, the Chebyshev--Tau and Chebyshev--Collocation methods, to solve for the atmospheric acoustic normal modes, and corresponding programs were developed. The two spectral methods successfully transform the problem of searching for the modal wavenumbers in the complex plane into a simple dense matrix eigenvalue problem by projecting the governing equation onto a set of orthogonal bases, which can be easily solved through linear algebra methods. After obtaining the eigenvalues and eigenvectors, the horizontal wavenumbers and their corresponding modes can be obtained with simple processing. Numerical experiments were examined for both downwind and upwind conditions to verify the effectiveness of the methods. The running time data indicated that both spectral methods proposed in this article are faster than the Legendre--Galerkin spectral method proposed previously.

cs.CE↗

arXiv2017

Improved Algorithm for Reconstructing Singular Connection in Multi-Block CFD Applications

An improved algorithm is proposed for the reconstruction of singular connectivity from the available pairwise connections during preprocessing phase. To evaluate the performance of the algorithm, an in-house computational fluid dynamics (CFD) code is used in which high-order finite-difference method for spatial discretization running on the Tianhe-1A supercomputer is employed. Test cases with a varied amount of mesh points are chosen, and the test results indicate that the improved singular connection reconstruction algorithm can achieve a speedup factor of 1000X or more when compared with the naive search method adopted in the former version of our code. Moreover, the parallel efficiency can benefit from the strategy of local communication based on the algorithm.

cs.CE↗

arXiv2017

Intrinsic mechanism for spectral evolution in single-frequency Raman fiber amplifier

In this work, the spectral evolution properties in single-frequency Raman fiber amplifier (RFA) with different pump manners are analyzed theoretically for the first time based on the gain dynamics. The analysis of gain dynamics reveals that the walk-off effect in counter-pumped manner produces a natural low-pass filter in single-frequency RFA. When applying rare-earth doped fiber lasers as the pump source, the strong temporal fluctuations in the pump source lead to spectral broadening in co-pumped manner, while the natural low-pass filter in countered-pumped case can still ensure single-frequency operation. Furthermore, applying temporal stable laser, such as single-frequency fiber laser or narrow band fiber laser spectral broadened by phase modulation technique, as the pump source would be superior for high-performance single-frequency RFA for the both two pump manners.

physics.optics↗

arXiv2015

Effects of temperature gradient on the interface microstructure and diffusion of diffusion couples: phase-field simulation

The temporal interface microstructure and diffusion in the diffusion couples with the mutual interactions of temperature gradient, concentration difference and initial aging time of the alloys were studied by phase-field simulation, the diffusion couples are produced by the initial aged spinodal alloys with different compositions. Temporal composition evolution and volume fraction of the separated phase indicates the element diffusion direction through the interface under the temperature gradient. The increased temperature gradient induces a wide single-phase region at two sides of the interface. The uphill diffusion proceeds through the interface, no matter the diffusion directions are up or down to the temperature gradient. For an alloy with short initial aging time, phase transformation accompanying the interdiffusion results in the straight interface with the single-phase regions at both sides. Comparing with the temperature gradient, composition difference of diffusion couple and initial aging time of the alloy show greater effect on the diffusion and interface microstructure.

cond-mat.mtrl-sci↗

arXiv2011

Cascades damage in α-iron with high damage energy

We present a numerical study of the cascades damage in α-iron for high cascade energy up to 250keV, corresponding to neutron energy above and beyond 14.1MeV. We observe that Frenkel pairs production efficiency and perfect crystal lattices damage efficiency have the close trend within the statistical uncertainty versus cascade energy, and the two kinds of curves first decrease with power-law trend, and arrive at their minimum at threshold energy of subcascades, then these curves begin to increase with cascade energy. The analysis demonstrates that the tendency of these curves reflects degree of concentration of defects and depends on the number of subcascades which keeps identical with defects scattering and the energy of subcascades reflecting defects concentration. The pair analysis technique may help us understand the total damage of perfect crystal lattices.

cond-mat.mtrl-sci↗

arXiv2007

Evaluation of Optical Magnitude of Deep Space Spacecraft

Optical-electric technology can measure the tangential position and velocity of spacecraft. To know the feasibility of the use of optical-electric technology, it is necessary to estimate the magnitude of spacecraft first. Since the spacecrafts are non-self-illumination objects, the estimation formulas of the optical magnitude of spacecraft is constructed according to the radiation theory and the extra-atmospheric radiant emittance of the Sun in the visible light wave band. Taking Chang'e-1 as an example, the magnitude of it in different situations is calculated.

astro-ph↗