A5100751850- Digital Media Forensic Detection
- Advanced Steganography and Watermarking Techniques
- QR Code Applications and Technologies
- Biometric Identification and Security
- Advanced Image and Video Retrieval Techniques
- Image Processing Techniques and Applications
- Face recognition and analysis
- Advanced Malware Detection Techniques
- Service-Oriented Architecture and Web Services
- IPv6, Mobility, Handover, Networks, Security
- Electromagnetic Compatibility and Noise Suppression
- Microwave Engineering and Waveguides
- Photonic and Optical Devices
- Hydrogels: synthesis, properties, applications
- Online Learning and Analytics
- Semiconductor Lasers and Optical Devices
- Handwritten Text Recognition Techniques
- Advanced Wireless Communication Techniques
- Caching and Content Delivery
- Adversarial Robustness in Machine Learning
- Network Packet Processing and Optimization
- Image Processing and 3D Reconstruction
- Antenna Design and Analysis
- Forensic and Genetic Research
- Online and Blended Learning
Shenzhen University
2016-2025
Tianjin Medical University
2024
Second Hospital of Tianjin Medical University
2024
Tianjin haihe hospital
2024
Shandong University
2023-2024
Shanghai Power Equipment Research Institute
2021-2024
System Equipment (China)
2024
Shenzhen Academy of Robotics
2019-2023
Hong Kong Polytechnic University
2023
Shandong Women’s University
2023
Face verification systems are prone to spoofing attacks on photos, videos, and 3D masks. detection, i.e., face anti-spoofing, liveness or presentation attack is an important task for securing in practice presents many challenges. In this paper, a state-of-the-art detection method based depth-based Fully Convolutional Network (FCN) revisited. Different supervision schemes, including global local label supervisions, comprehensively investigated. A generic theoretical analysis associated...
Epitaxial growth is one of the most commonly used strategies to precisely tailor heterostructures with well-defined compositions, morphologies, crystal phases, and interfaces for various applications. However, as epitaxial requires a small interfacial lattice mismatch between components, it remains challenge synthesis constructed by materials large and/or different chemical bonding, especially noble metal-semiconductor heterostructures. Here, we develop metal-seeded strategy prepare highly...
Nowadays, 2D barcodes have been widely used as an interface to connect potential customers and advertisement contents. However, the appearance of a conventional barcode pattern is often too obtrusive for integrating into aesthetically designed advertisement. Besides, no human readable information provided before successfully decoded. This paper proposes new picture-embedding barcode, called PiCode, which mitigates these two limitations by equipping scannable with picturesque appearance....
Gold nanoparticles functionalized with pro-apoptotic peptide (PAP-AuNPs) were fabricated, which able to lead programmed cell-death by damaging mitochondria.
Two kinds of representative polymers, poly(N-isopropylacrylamide) (PNIPAAm) and β-cyclodextrin (β-CD) were selected modified with azide alkyne fucntional groups, respectively. When the solutions these two polymers mixed together, a cross-linking reaction, type Huisgen's 1,3-dipolar azide-alkyne cycloaddition, occurred in presence Cu(I) catalyst. The strategy described here provides several advantages for hydrogel formation including mild reaction conditions controllable gelation rate....
The finite rate of innovation (FRI) principle is developed for sampling a class non-bandlimited signals that have number degrees freedom per unit time, i.e., with FRI. This scheme later extended to three classes kernels compact support and applied the step edge reconstruction problem by treating image row row. In this paper, we regard edges as 2D FRI reconstruct them block block. parameters are obtained from moments given Experimentally, our technique can more precisely track Cramér-Rao...
Compared with conventional authentication schemes, which utilize cryptographic mechanisms and operate at an upper layer, the physical (PHY) layer possesses many advantages, such as, higher information-theoretic security by introducing uncertainty to adversary, better efficiency, compatibility avoiding any operations layer. An effective PHY-layer scheme should consider covertness, security, robustness together. However, in published literature, these three properties have been separately...
The traditional two-dimensional (2D) barcode has been employed in anti-counterfeiting systems as a storage media for serial numbers. However, an attack can be initiated by simply copying the 2D and attaching it to counterfeit product. In this paper, we aim at proposing authentication scheme with mobile imaging device barcode. This work presents competitive solution among schemes that have verified under conditions. proposed copy-proof is composed of two sets features which are extracted...
Inspired by the philosophy employed human beings to determine whether a presented face example is genuine or not, i.e., glance at globally first and then carefully observe local regions gain more discriminative information, for anti-spoofing problem, we propose novel framework based on Convolutional Neural Network (CNN) Recurrent (RNN). In particular, model behavior of exploring face-spoofing-related information from image sub-patches leveraging deep reinforcement learning. We further...
Recapturing attack can be employed as a simple but effective anti-forensic tool for digital document images. Inspired by the inspection process that compares questioned against some known samples, we proposed recapture detection scheme employing Siamese network to compare and extract distinct features in recaptured image. The algorithm takes advantage of both metric learning image forensic techniques, forms triplets considering important factors authentication, e.g., types, resolutions,...
Recently, some picture-embedding schemes have been proposed to improve the aesthetic appearance of 2D barcodes. However, these barcodes are not robust distortions incurred by print/display-and-capture channel under limited rendering space and resolution. In this paper, a barcode named Robust Aesthetic (RA) Code is counter impairments, such as downsampling error inter-symbol interference. Experimental results demonstrate that RA more than existing barcodes, especially printing/displaying...
Document recapturing is a presentation attack that covers the forensic traces in digital domain. detection (DPAD) an important step document authentication pipeline. Existing DPAD methods suffer from low generalization performance under cross-domain scenario with different types of documents. Data augmentation de facto technique to reduce risk overfitting training data and improve generalizability trained model. In this work, we approaches by addressing two limitations existing frequency...
High-entropy alloy (HEA) nanoparticles (NPs) have been emerging with superior compositional tunability and multielemental synergy, presenting a unique platform for material discovery performance optimization. Here we report synthetic approach utilizing hollow-carbon confinement in the ordinary furnace annealing to achieve nonequilibrium HEA-NPs such as Pt0.45Fe0.18Co0.12Ni0.15Mn0.10 uniform size ∼5.9 nm. The facile temperature control allows us not only reveal detailed reaction pathway...
Recent advances in deep learning algorithms have shown impressive progress image copy-move forgery detection (CMFD). However, these lack generalizability practical scenarios where the copied regions are not present training images, or cloned part of background. Additionally, utilize convolution operations to distinguish source and target regions, leading unsatisfactory results when blend well with To address limitations, this study proposes a novel end-to-end CMFD framework that integrates...