A5005658072- Biometric Identification and Security
- Advanced Vision and Imaging
- Image Processing Techniques and Applications
- Forensic Fingerprint Detection Methods
- Advanced Steganography and Watermarking Techniques
- Advanced Image Processing Techniques
- Cell Image Analysis Techniques
- Brain Tumor Detection and Classification
- Digital Media Forensic Detection
- Network Security and Intrusion Detection
- User Authentication and Security Systems
- Internet Traffic Analysis and Secure E-voting
- Geotechnical Engineering and Analysis
- Face and Expression Recognition
- Geomechanics and Mining Engineering
- Grouting, Rheology, and Soil Mechanics
- Geoscience and Mining Technology
- Coal Properties and Utilization
- Gait Recognition and Analysis
- Digital Imaging in Medicine
- AI in cancer detection
- Advanced Neural Network Applications
- Target Tracking and Data Fusion in Sensor Networks
- Medical Imaging and Analysis
- Smart Grid Security and Resilience
Xi'an Jiaotong University
2024
Griffith University
2024
University of Canberra
2016-2023
UNSW Sydney
2016-2023
Nanjing University of Information Science and Technology
2023
ORCID
2021
Chang'an University
2018-2019
Meitan General Hospital
2016
Tianjin University
2013-2015
Abstract The Cell Tracking Challenge is an ongoing benchmarking initiative that has become a reference in cell segmentation and tracking algorithm development. Here, we present significant number of improvements introduced the challenge since our 2017 report. These include creation new segmentation-only benchmark, enrichment dataset repository with datasets increase its diversity complexity, silver standard corpus based on most competitive results, which will be particular interest for...
Latent fingerprint enhancement is an essential preprocessing step for latent identification.Most methods try to restore corrupted gray ridges/valleys.In this paper, we propose a new method that formulates as constrained generation problem within generative adversarial network (GAN) framework.We name the proposed FingerGAN.It can enforce its generated (i.e, enhanced fingerprint) indistinguishable from corresponding ground truth instance in terms of skeleton map weighted by minutia locations...
False data injection attacks (FDIAs) have recently become a major threat to smart grids. Most of the existing FDIA detection methods focused on modeling temporal relationship time-series measurement but paid less attention spatial between bus/line and failed consider subgrids. To address these issues, in this article, we propose subgrid-oriented microservice framework by integrating well-designed spatial–temporal neural network for ac-model power systems. First, is developed model...
Contactless fingerprint recognition is highly promising and an essential component in the automatic identification system. However, due to inherent characteristic of perspective distortions contactless fingerprints, achieving a accurate system very challenging. In this paper, we propose robust method based on global minutia topology loose genetic algorithm. order avoid inaccurate minutiae alignment problem suffered conventional transformation-based methods, correspondence established by...
3D Gaussian Splatting has emerged as one of the most prominent algorithms for novel view synthesis in recent years, with numerous studies adapting it to dynamic scenes, such employing deformation MLP predict motion. However, existing methods frequently overlook contextual information within temporal sequences, resulting inaccuracies motion modeling. To mitigate this issue, we propose KF-GS, which is inspired by Kalman filter algorithm and integrates both observation prediction estimate...
An automated fingerprint recognition system (AFRS) for 3D fingerprints is essential and highly promising biometric security. Despite the progress in developing AFRSs, achieving high-quality real-time reconstruction high-accuracy of remain two challenging issues. To address them, we propose a robust AFRS based on ridge-valley (RV)-guided topology polymer (TTP) feature extraction. The former considers unique characteristics RV achieves reconstruction. Unlike traditional triangulation-based...
Contactless fingerprint biometrics has achieved rapid development in the past decades thanks to its inherent advantages, such as no physical contact between a finger and sensor, contamination by latent fingerprints, more hygienic. These advantages have paved way for new 2D or 3D contactless fingerprint-based applications promoted larger number of academic publications recent years. Therefore, it is necessary important conduct comprehensive survey on biometric technology, review latest...
Latent fingerprints are important evidences used by law enforcement agencies to identify suspects for centuries. However, due the poor image quality and complex background noise, separating fingerprint region-of-interest from is a very challenging problem. This paper proposes new latent segmentation method based on Convolutional Neural Networks (ConvNets). The problem formulated as classification system, in which set of elaborately designed ConvNets learned classify each patch either or...
Abstract Motivation Live cell segmentation is a crucial step in biological image analysis and also challenging task because time-lapse microscopy sequences usually exhibit complex spatial structures complicated temporal behaviors. In recent years, numerous deep learning-based methods have been proposed to tackle this obtained promising results. However, designing network with excellent performance requires professional knowledge expertise very time-consuming labor-intensive. Recently emerged...
Fingerprint authentication techniques have been employed in various Internet of Things (IoT) applications for access control to protect private data, but raw fingerprint template leakage unprotected IoT may render the system insecure. Cancelable templates can effectively prevent privacy breaches and provide strong protection original templates. However, suit resource-constrained devices, oversimplified would compromise performance significantly. In addition, length existing cancelable is...
Smart grids are vulnerable to stealthy false data injection attacks (SFDIAs), as SFDIAs can bypass residual-based bad detection mechanisms. Methods based on deep learning technology have shown promising accuracy in the of SFDIAs. However, most existing methods rely temporal structure a sequence measurements but do not take account spatial between buses and transmission lines. To address this issue, we propose spatiotemporal network, PowerFDNet, for SFDIA AC-model power grids. The PowerFDNet...
Segmentation of white matter tracts in diffusion magnetic resonance images is an important first step many imaging studies the brain health and disease. Similar to medical image segmentation general, a popular approach tract use U-Net based artificial neural network architectures. Despite suggested improvements architecture recent years, there lack systematic comparison architectural variants for segmentation. In this paper, we evaluate multiple architectures specifically purpose. We compare...
Microscopy cell segmentation is a crucial step in biological image analysis and challenging task. In recent years, deep learning has been widely used to tackle this task, with promising results. A critical aspect of training complex neural networks for purpose the selection loss function, as it affects process. field segmentation, most research improving function focuses on addressing problem inter-class imbalance. Despite achievements, more work needed, challenge not only imbalance but also...