- Physical Unclonable Functions (PUFs) and Hardware Security
- E-commerce and Technology Innovations
- Anomaly Detection Techniques and Applications
- Adversarial Robustness in Machine Learning
- Bacillus and Francisella bacterial research
- Supply Chain Resilience and Risk Management
- Advanced Technology in Applications
- Infrastructure Resilience and Vulnerability Analysis
- Neural Networks and Applications
Dalian University of Technology
2024
Northwest Institute of Mechanical and Electrical Engineering
2021-2023
The collective Unmanned Weapon System-of-Systems (UWSOS) network represents a fundamental element in modern warfare, characterized by diverse array of unmanned combat platforms interconnected through heterogeneous architectures. Despite its strategic importance, the UWSOS is highly susceptible to hostile infiltrations, which significantly impede battlefield recovery capabilities. Existing methods enhance resilience predominantly focus on basic graph relationships, neglecting crucial...
In the robustness verification of neural networks, formal methods have been used to give deterministic guarantees for networks. However, recent studies found that method single-neuron relaxation in this field has an inherent convex barrier affects its capability. To address problem, we propose a new by combining dual-neuron and linear programming. This captures dependencies between different neurons same hidden layer adding two-neuron joint constraint programming model, thus overcoming...
With the acceleration of intelligent process domestic industry, manipulator in production workshop is increasingly widely used, and collision problem becoming more prominent. Aiming at this problem, AABB hierarchical bounding box algorithm designed based on algorithm. The commonly used are to envelope human 3D point cloud data model. results show that operation time 18.11% 25.04% shorter than OBB k-DOPs algorithms, but proportion body volume only reduced by 6.65% 11.20%, comprehensive...
In harsh environments such as rain, snow, fog, and haze, the target perception capability of optoelectronic system is seriously reduced. At same time, embedded image processing hardware platform completes high-definition video preprocessing, detection other tasks with slow response time-consuming. The key frame decoding proposed in this paper reduces requirement tracker on computing power system, enhancement influence environment tracking effect. problem converted into a...