Peng Wu

ORCID: 0000-0003-4784-8984
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Advanced Neural Network Applications
  • Software Testing and Debugging Techniques
  • Formal Methods in Verification
  • Advancements in Battery Materials
  • Advanced Malware Detection Techniques
  • Adversarial Robustness in Machine Learning
  • Software Reliability and Analysis Research
  • Model-Driven Software Engineering Techniques
  • Topology Optimization in Engineering
  • Visual Attention and Saliency Detection
  • Evaluation and Optimization Models
  • Electromagnetic wave absorption materials
  • Electrical Fault Detection and Protection
  • Optical Imaging and Spectroscopy Techniques
  • Conducting polymers and applications
  • Remote-Sensing Image Classification
  • Advanced Image and Video Retrieval Techniques
  • Non-Invasive Vital Sign Monitoring
  • Artificial Immune Systems Applications
  • Advanced Computational Techniques and Applications
  • Power Systems Fault Detection
  • Advanced Battery Materials and Technologies
  • Heart Rate Variability and Autonomic Control
  • Peer-to-Peer Network Technologies
  • Textile materials and evaluations

Beijing Jiaotong University
2014-2024

Zhejiang Sci-Tech University
2023-2024

Zhong Ke San Huan (China)
2024

Guizhou Electric Power Design and Research Institute
2023

Shanxi University
2023

Taiyuan Institute of Technology
2022

Mälardalen University
2021

Institute of Software
2020

Chinese Academy of Sciences
2020

University of Chinese Academy of Sciences
2020

In unmanned aerial vehicle photographs, object detection algorithms encounter challenges in enhancing both speed and accuracy for objects of different sizes, primarily due to complex backgrounds small objects. This study introduces the PDWT-YOLO algorithm, based on YOLOv7-tiny model, improve effectiveness across all sizes. The proposed method enhances by incorporating a dedicated small-object layer, while reducing conflict between classification regression tasks through replacement model’s...

10.3390/a16110520 article EN cc-by Algorithms 2023-11-14

With the exponential growth of remote sensing images in recent years, there has been a significant increase demand for micro-target detection. Recently, effective detection methods small targets have emerged; however, micro-targets (even fewer pixels than targets), most existing are not fully competent feature extraction, target positioning, and rapid classification. This study proposes an enhanced method, especially micro-targets, which combined loss function (consisting NWD CIOU) is used...

10.3390/info15020108 article EN cc-by Information 2024-02-12

Urgent natural environmental events, such as floods, power failures, and epidemics, result in disruptions to the traffic system heavy disturbances public requirements. In order strengthen ability of transport network handle urgent issues, this paper simulates disruption situation stations urban agglomeration by attacking nodes, evaluates resist (i.e., invulnerability). Firstly, model integrated passenger is established based on complex theory. The highway network, railway coupling are...

10.3390/ijerph20010450 article EN International Journal of Environmental Research and Public Health 2022-12-27

Metamorphic testing (MT) is an innovative approach to alleviating the oracle problem in software testing, which uses metamorphic relations of program under test, instead test oracles, verify its outputs. To alleviate web services, we had previously proposed MT framework for services. In this paper, further improve efficiency and automation by leveraging iteratively generate cases. We present a fixed-size iterative algorithm implement it framework. conduct three case studies evaluate fault...

10.1504/ijwgs.2020.110945 article EN International Journal of Web and Grid Services 2020-01-01

Remote photoplethysmography (rPPG) refers to a non-contact technique that measures heart rate through analyzing the subtle signal changes of facial blood flow captured by video sensors. It is widely used in contactless medical monitoring, remote health management, and activity providing more convenient non-invasive way monitor health. However, factors such as ambient light variations, movements, differences absorption reflection pose challenges deep learning-based methods. To solve these...

10.3390/s24216791 article EN cc-by Sensors 2024-10-22

Due to their high reliability, composite materials are commonly used in industry. The stacking sequence is crucial the strength of materials; however, arrangement this very time consuming. Because usually NP complete, genetic algorithms (GA) were employed deal with problems study. interlaminate stress σ33 a constant thickness plate, which obtained by analytical method, was as fitness function GA find its minimum value arrange ply sequence. Three examples and symmetrical laminate plates...

10.1080/02286203.2006.11442352 article EN International Journal of Modelling and Simulation 2006-01-01

The access of distributed power sources to distribution networks not only improves the reliability supply, but also has a negative impact on relay protection schemes. This brought difficulty fault detection and significant workload search for points, resulting in prolonged outages damage electrical equipment, seriously affecting supply reliability. paper proposes novel diagnosis algorithm network lines based multivariate data fusion. location equation is derived from matrix inversion lemma,...

10.1109/peeec60561.2023.00179 article EN 2023-09-25

In recent years, the study of adversarial robustness in object detection systems, particularly those based on deep neural networks (DNNs), has become a pivotal area research. Traditional physical attacks targeting detectors, such as patches and texture manipulations, directly manipulate surface object. While these methods are effective, their overt manipulation objects may draw attention real-world applications. To address this, this paper introduces more subtle approach: an inconspicuous...

10.48550/arxiv.2410.10091 preprint EN arXiv (Cornell University) 2024-10-13

Decentralized power systems are commonly used in high-speed trains. However, many parameters decentralized uncertain and inevitably have errors. We present a reasoning method based on the interval numbers for Uncertain their unavoidable errors quantitatively described by numbers. also define generalized linear equations with (LAIs), which can be to describe movement of train. Furthermore, it is proven that zero sets LAIs convex. Therefore, inside fault-tolerance area formed vertexes edges...

10.1155/2021/6624528 article EN Mathematical Problems in Engineering 2021-01-04

In this paper, CoFe2O4@pyrolyzed bacterial cellulose (CFO@PBC) nanocomposites were synthesized by a hydrothermal process and an in situ composite technique as the cathode for rechargeable Li-O2 batteries (LOB). CoFe2O4@Ketjenblack (CFO@KB) prepared same way comparison. The as-prepared materials characterized SEM, XRD, BET. LOB with CFO@PBC-3 showed maximum initial discharge capacity of 2660 mAh g−1 (electrode) at current density 100 mAg−1, while mass ratio CoFe2O4 pyrolyzed was 3. cyclic...

10.3390/cryst12081165 article EN cc-by Crystals 2022-08-19

Herein, a three-step approach toward multi-layered porous PBC/graphene sandwich has been developed, in which the chemical bonding interactions have successfully enhanced via esterification between layers of pyrolyzed bacterial cellulose (PBC) and graphene. Such chemically induced compatible interface demonstrated to contribute significantly mass transfer efficiency when is deployed as electrode material for both supercapacitors lithium-sulfur batteries. The high specific capacitance...

10.3390/ma15196709 article EN Materials 2022-09-27

Error coefficients are ubiquitous in systems. In particular, errors reasoning verification must be considered regarding safety-critical We present a method that can applied to systems described by the polynomial error assertion (PEA). The implication relationship between PEAs converted an inclusion zero sets of PEAs; then transformed into first-order logic. Combined with quantifier elimination method, based on cylindrical algebraic decomposition, judgment is parameters and specific...

10.3390/info12080309 article EN cc-by Information 2021-07-30

Firstly, this paper analyzes three forms of profit distribution based on two-level capacity cooperation service supply chain between supplier and integrator: fixed payment pattern, mixed pattern share pattern; points out that is the development trend supplier. Then, proposes principles enterprises should follow during distribution, including consistent incomes contribution, symmetrical risk benefit weighted principle. it integrator may impact proportion pertaining to quality control...

10.1109/icsssm.2014.6943371 article EN 2014-06-01

10.3724/sp.j.1087.2013.00323 article EN Journal of Computer Applications 2013-09-23

10.1109/iceict57916.2023.10245294 article EN 2022 IEEE 5th International Conference on Electronic Information and Communication Technology (ICEICT) 2023-07-21

The quorum generation algorithm proposed in this paper is based on the quadratic difference set and initializes with a sequence of prime numbers. Initialization numbers increases fairness marking corresponding elements D set. This makes our slightly better than time complexity O(N), still O(N). size generated close to 3N, when number nodes 10 million.

10.1145/3425329.3425345 article EN 2020-09-25

Error parameters are inevitable in systems. In formal verification, previous reasoning methods seldom considered the probability information of errors. this article, errors described as symmetric truncated normal intervals consisting and density. Furthermore, we also rigorously prove lemmas a theorem to partially simplify calculation process independently verify formulas variance expectation interval given by some scholars. The mathematical derivation or verification codes provided for most...

10.3390/sym14010025 article EN Symmetry 2021-12-25
Coming Soon ...