A5100736939- Video Surveillance and Tracking Methods
- Infrared Target Detection Methodologies
- Advanced MIMO Systems Optimization
- 3D Shape Modeling and Analysis
- Advanced Vision and Imaging
- Chaos-based Image/Signal Encryption
- Computer Graphics and Visualization Techniques
- Advanced Photocatalysis Techniques
- Environmental remediation with nanomaterials
- Quantum Dots Synthesis And Properties
- Digital Rights Management and Security
- Green IT and Sustainability
- Microgrid Control and Optimization
- Biometric Identification and Security
- Neural Networks Stability and Synchronization
- ECG Monitoring and Analysis
- Digital Media Forensic Detection
- Remote Sensing in Agriculture
- Internet Traffic Analysis and Secure E-voting
- Gear and Bearing Dynamics Analysis
- Numerical Methods and Algorithms
- Generative Adversarial Networks and Image Synthesis
- Random lasers and scattering media
- Power Systems Fault Detection
- EEG and Brain-Computer Interfaces
Anhui Science and Technology University
2025
Anhui University of Science and Technology
2025
Chinese People's Armed Police Force Engineering University
2024
Chongqing University of Posts and Telecommunications
2024
Wuhan University
2023-2024
Chinese University of Hong Kong
2024
Huaqiao University
2024
Northwest Normal University
2024
Southwest Minzu University
2022
China Southern Power Grid (China)
2019
A simple zinc salt additive strategy to improve the quality of CdS and Sb 2 S 3 films for photovoltaic application.
Pseudo-random number generators (PRNGs) are deterministic algorithms that generate sequences of numbers approximating the properties random numbers, which widely utilized in various fields. In this paper, we present a Genetic Algorithm Optimized Generative Adversarial Network (hereinafter referred to as GAGAN), is designed for effective training discrete generative adversarial networks. situations where non-differentiable activation functions, such modulo operation, employed and traditional...
The interaction between the converter controllers of doubly-fed induction generators (DFIGs) and fixed series compensation tends to cause a new type subsynchronous resonance (SSR), or control interaction. To address this issue, paper proposes novel scheme deactivate such an by embedding suppression filters (SSFs) into DFIG's controllers. Its ability stabilize SSR is validated applying it practical wind farm system in North China, which has experienced many events. electromagnetic model...
The impulse features are typical characteristics in vibration signals when rotating equipment failures occur, such as faults of bearings and gears. Accurate capture can help better monitor the operating status equipment. As an energy-concentrated time-frequency analysis (TFA) tool, time-reassigned synchrosqueezing transform (TSST) is effective means to characterize signals. However, existence some obvious defects limits its application. For example, strong frequency-varying cannot be...
As Large Language Models (LLMs) gain great success in real-world applications, an increasing number of users are seeking to develop and deploy their customized LLMs through cloud services. Nonetheless, some specific domains, there still concerns regarding cost trade-offs between privacy issues accuracy. In this study, we introduce a cost-effective self-adaptive LLM shaking tuning recovery mechanism, named CypherTalk. With carefully designed horizontal vertical operators, can achieve...
In mobile social networks, users can easily communicate with others through smart devices. Therefore, the protection of user privacy in networks is becoming a significant subject. To solve this problem, paper proposes fine-grained data access control scheme that uses attributes to match friends. our scheme, friend-making parties generate friend preference and self-description lists, respectively, realizing attribute hiding by converting friendship into ciphertext policies keys. The platform...
<title>Abstract</title> The application of infrared (IR) spectroscopy is limited due to intrinsically theoretical constraints. Machine learning offers a route for overcoming these limitations. Nevertheless, its efficacy significantly relies on the quality data because irrelevant and/or redundant information hinders success true signal. In this study, we introduced an intuitive multivariate method tailored assessing spectra obtained from experimental replicates identical samples....
Recent text-to-3D generation methods achieve impressive 3D content creation capacity thanks to the advances in image diffusion models and optimizing strategies. However, current struggle generate correct for a complex prompt semantics, i.e., describing multiple interacted objects binding with different attributes. In this work, we propose general framework named Progressive3D, which decomposes entire into series of locally progressive editing steps create precise prompts, constrain change...
Artificial Intelligence (AI) has achieved significant advancements in technology and research with the development over several decades, is widely used many areas including computing vision, natural language processing, time-series analysis, speech synthesis, etc. During age of deep learning, especially arise Large Language Models, a large majority researchers' attention paid on pursuing new state-of-the-art (SOTA) results, resulting ever increasing model size computational complexity. The...
Cellular controlled Device-to-Device (D2D) system is a local service for future communication systems like LTE, LTE-A, etc. D2D communications can share the Uplink (UL) or Downlink (DL) resources with cellular to improve throughput. Since reusing UL resource DL has different merits in scenarios, if only reuses either during whole procedure, there may be some problems, such as load balance, interference, This paper proposes method adaptively TDD UL/DL slot utilization according change of...
Adversarial camouflage is a widely used physical attack against vehicle detectors for its superiority in multi-view performance. One promising approach involves using differentiable neural renderers to facilitate adversarial optimization through gradient back-propagation. However, existing methods often struggle capture environmental characteristics during the rendering process or produce textures that can precisely map target vehicle, resulting suboptimal Moreover, these approaches neglect...
3D generation has witnessed significant advancements, yet efficiently producing high-quality assets from a single image remains challenging. In this paper, we present triplane autoencoder, which encodes models into compact latent space to effectively compress both the geometry and texture information. Within autoencoder framework, introduce 3D-aware cross-attention mechanism, utilizes low-resolution representations query features high-resolution feature volume, thereby enhancing...
Synchronization on multi-layer networks are paid increasingly attention in recent years, and the interplay between network dynamics structures remains open. Focusing interactions synchronization from perspective of coupled delays, this paper investigates intra-layer inter-layer delayed duplex without with additional controllers via Lyapunov stability theory Halanay inequality, presents criteria which not only applied for general topologies but case coupling gives upper bound delay thresholds...
Prior works on physical adversarial camouflage against vehicle detectors mainly focus the effectiveness and robustness of attack. The current most successful methods optimize 3D texture at a pixel level. However, this results in conspicuous attention-grabbing patterns generated camouflage, which humans can easily identify. To address issue, we propose Customizable Natural Camouflage Attack (CNCA) method by leveraging an off-the-shelf pre-trained diffusion model. By sampling optimal image...
Adversarial camouflage is a widely used physical attack against vehicle detectors for its superiority in multi-view performance. One promising approach involves using differentiable neural renderers to facilitate adversarial optimization through gradient back-propagation. However, existing methods often struggle capture environmental characteristics during the rendering process or produce textures that can precisely map target vehicle. Moreover, these approaches neglect diverse weather...
On account of the fact that conventional synchrophasor estimators are difficult to meet requirements accuracy and rapidity at same time, a comprehensive adaptive estimator is proposed in this paper on rapidity, accuracy, reliability expansibility. In order verify effectiveness estimator, The simulation test analysis with carried out MATLAB/Simulink platform. Moreover, RT-LAB platform for wide area measurement control distribution network built hardware-in-loop tests. A typical case based...
The implementation of FINFET devices in the SRAM cell provides many benefits over that planar bulk devices. short channel effect, drive current and mismatch can be better controlled. Several FIN number options among PU(pull up device), PD(pull down device) PG(pass gate selected to achieve good read noise margin write margin. But highest-density cell, order minimize bit-cell area, 3 are designed as one only for each. is suffered deeply by this option unfortunately Some techniques presented...
During the process of Electrocardiogram (ECG) recognition, frequent challenges arise concerning lack convenience and accuracy. Moreover, imbalanced distribution ECG data presents significant hurdles for recognition models. To address aforementioned issues, paper proposes two models on a dataset enhanced through Generative Adversarial Network (GAN): XGVoting CRANet, which are specifically designed to tackle lightweight design accuracy improvement. By utilizing GAN augmentation, is effectively...
To increase the traction force of telescopic downhole robot, support mechanism robot was optimized, and a slider-spring transmission introduced to solve problem jamming. A mathematical model established, covering relationship between contact normal pressure hydraulic cylinder load. The finite element simulation friction block established based on Ansys software. Stress analysis casing under different pressures carried out, frictional characteristics were obtained. results show that composed...
The paper is based on PHM oriented design and modeling of durability testing for clamping devices. scheme made out through the Computer Aided Engineering (CAE) simulation result degradation model studied by Data-driven according to test data. Based analysis main failure mode mechanism, it concluded that reason device cracking fatigue which caused cycle force. conducted CAE analysis, strain data have been collected NI acquisition equipment. data, method used high-cycle And with comparison...