A5101973599- Chaos-based Image/Signal Encryption
- Cryptographic Implementations and Security
- Complex Network Analysis Techniques
- Physical Unclonable Functions (PUFs) and Hardware Security
- Advanced Graph Neural Networks
- Advanced Vision and Imaging
- Privacy-Preserving Technologies in Data
- Robotics and Sensor-Based Localization
- Network Security and Intrusion Detection
- Video Surveillance and Tracking Methods
- Graph Theory and Algorithms
- Zebrafish Biomedical Research Applications
- Radiomics and Machine Learning in Medical Imaging
- COVID-19 diagnosis using AI
- Smart Grid Security and Resilience
- Cholesterol and Lipid Metabolism
- Game Theory and Applications
- Cloud Computing and Resource Management
- Distributed and Parallel Computing Systems
- Bioinformatics and Genomic Networks
- Automated Road and Building Extraction
- Adversarial Robustness in Machine Learning
- Information and Cyber Security
- Electrostatic Discharge in Electronics
- Robotic Path Planning Algorithms
Hunan University of Finance and Economics
2020-2025
Hunan University
2013-2024
Changchun University of Technology
2022
Erasmus University Rotterdam
2020-2021
Erasmus MC
2020-2021
Nanjing University of Information Science and Technology
2018-2019
Kunming University of Science and Technology
2011
Animal pose estimation tools based on deep learning have greatly improved animal behaviour quantification. These perform individual video frames, but do not account for variability of body shape in their prediction and evaluation. Here, we introduce a novel multi-frame framework, referred to as OptiFlex. This framework integrates flexible base model (i.e., FlexibleBaseline), which accounts shape, with an OpticalFlow that incorporates temporal context from nearby frames. Pose can be optimised...
Mobile Edge Computing (MEC) has been successfully applied to smart homes, mobile devices, and other electronic products, offering reduced communication latency alleviating pressure on cloud services. The integration of Federated Learning (FL) MEC provides customized services for enhancing user experiences. However, there are still gaps in combining these technologies: FL is susceptible membership inference attacks, risking privacy; faces limitations energy computational resources,...
Abstract Deep learning based animal pose estimation tools have greatly improved behaviour quantification. However, those all make predictions on individual video frames and do not account for variability of body shape in their model designs. Here, we introduce the first video-based architecture, referred to as OptiFlex, which integrates a flexible base with an optical flow incorporate temporal context from nearby frames. This approach can be combined multi-view information, generating...
Building the security mechanism for Controller Area Network (CAN) to defend against attack has drawn substantial attention recently. Fingerprinting ECUs provide ability of authentication based on physical characteristics can protect CAN network effectively. The clock skew which is unique and stable be exploited pinpoint attacker detect intrusion. However, a common downside existing clock-skew-based approaches that estimation process affected by message scheduling or arbitration. In our work,...
One area of current research on adversarial attacks is how to generate plausible examples when only a small number datasets are available. Current attack algorithms used these black-box systems face challenges, such as difficulty in training convergence, ambiguous sample images, substitute models collapse, unsatisfactory success rates, high query cost, and low defense capability improvement target models. As result, constructing situations few known real-world circumstances remains...
With the fast development of social networks, more and data has been generated. Finding useful information among these is important. Subgraph matching method that can be used in networks for search, recommender systems, so on. In this paper, we choose five typical graph methods which are VF2, SPath, Turbo <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ISO</sub> , BoostIso, RI to value their performance scalability networks. These verified by...
Modular exponentiation, an operation widely utilized in cryptographic protocols to transfer text and other forms of data, can also be applied Internet-of-Things (IoT) devices with high security requirements. However, due the resource consumption modular IoT face problem insufficient. Fortunately, secure outsourcing scheme offers a new solution for resource-constrained devices. In this article, we apply parallel provide possibility exponentiation operation, which is used After that, task...
Most unsupervised methods of person re-identification (Re-ID) obtain pseudo-labels through clustering. However, in the process clustering, hard quantization loss caused by clustering errors will make model produce false pseudo-labels. In order to solve this problem, an based on softened labels training method is proposed. The innovation that correlation among image features used find reliable positive samples and train them a smooth manner. To further explore features, some modules are...
Summary With the rise of industrialization, importance industrial Internet Things (IIoT) has increased significantly, and with it comes a variety security threats. Therefore, these networks is critical. Industrial Response Systems (IRSs), as last line security, plays an important role in system Things. In this paper, new IRS model based on non‐cooperative game proposed. First, by combining Partially Observable Markov Decision Process (POMDP) stochastic expanded attack tree, our could...
Learning discriminative features for facial expression recognition (FER) in the wild is a challenging task due to significant intra-class variations, inter-class similarities, and extreme class imbalances. In order solve these issues, contrastive-learning-based extra-contrast affinity network (ECAN) method proposed. The ECAN consists of feature processing two proposed loss functions, namely extra negative supervised contrastive (ENSC loss) multi-view (MVA loss). provides current historical...
SM4 is a block cipher proposed by the Chinese government. Strengthening research and extension of significant to development promotion cryptography standards. To date, relevant rare. Thus, we propose implementation an algorithm resistant t o power analysis. Ideally, secure masking scheme used for cipher, which particularly suited in application specific integrated circuit. Moreover, mask our chip process improved make safer. Simulation results confirm that use counteractive measures analysis...
An active shield is a sensor mesh devised and implemented on top of most secure chips or in embedded system as good protection for circuits sensitive data stored the chips. It one effective annoyances microprobing attackers. But, rerouting attack can defeat by disclosing geometry structure shield, causing short circuit mesh. This paper proposes dynamic to resist attack. A which contains some routing nodes connects metal lines between adjacent nodes. Changing connection statuses results...
In this paper, we apply a protective strategy to the AES encryption algorithm. order reduce power consumption and correlation in process of arithmetic operations, change trigger number location hardware implementation by modifying structure Through simulation platform, our Advanced Encryption Standard algorithm can effectively resist Differential Power Analysis Correlation analysis. Compared with mask dynamic differential logic, policy doesn't increase complexity area.
Functional verification of large scale circuit design is a basic problem in Very Large Scale Integrated (VLSI) design. With the increasing circuit, it urgent to divide whole into some smaller sub-circuits so as perform parallel functional on multiple hardware processors. The partition hardware-accelerated can be regarded graph problem. However, unlike traditional requirements for minimum cutting, needs reduce simulation depth and improve parallelism simulation. Therefore, combined with...