Dinh Cong Nguyen

ORCID: 0000-0003-0798-5511
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Privacy-Preserving Technologies in Data
  • Advanced Image and Video Retrieval Techniques
  • Handwritten Text Recognition Techniques
  • Advanced Neural Network Applications
  • Image Retrieval and Classification Techniques
  • Vehicle License Plate Recognition
  • Smart Agriculture and AI
  • Spectral Theory in Mathematical Physics
  • Advanced MIMO Systems Optimization
  • Power Line Inspection Robots
  • UAV Applications and Optimization
  • Wireless Communication Security Techniques
  • Colorectal Cancer Screening and Detection
  • Context-Aware Activity Recognition Systems
  • Advanced Differential Equations and Dynamical Systems
  • Cooperative Communication and Network Coding
  • Multimodal Machine Learning Applications
  • Vehicular Ad Hoc Networks (VANETs)
  • Advanced Topics in Algebra
  • Digital Media Forensic Detection
  • Smart Parking Systems Research
  • Digital Transformation in Industry
  • Underwater Vehicles and Communication Systems
  • Video Surveillance and Tracking Methods
  • Water Quality Monitoring Technologies

University of Alabama in Huntsville
2024-2025

Hồng Đức University
2019-2024

Université de Tours
2019-2020

Image and Pervasive Access Laboratory
2017

Sorbonne Université
2017

10.1016/j.bspc.2024.105979 article EN Biomedical Signal Processing and Control 2024-02-08

Accurate perception is essential for advancing autonomous driving and addressing safety challenges in modern transportation systems. Despite significant advancements computer vision object recognition, current methods still face difficulties complex real-world traffic environments. Challenges such as physical occlusion limited sensor field of view persist individual vehicle Cooperative Perception (CP) with Vehicle-to-Everything (V2X) technologies has emerged a solution to overcome these...

10.48550/arxiv.2310.03525 preprint EN cc-by-sa arXiv (Cornell University) 2023-01-01

10.1515/rose.1999.7.4.303 article EN Random Operators and Stochastic Equations 1999-01-01

We prove that the set of linear cocycles with simple Lyapunov spectrum is dense in space all equipped uniform topology.

10.1017/s014338579915199x article EN Ergodic Theory and Dynamical Systems 1999-12-01

The spectral theory of the one-dimensional Schrödinger operator with a quasi-periodic potential can be fruitfully studied considering corresponding differential system. In fact presence an exponential dichotomy for system is equivalent to statement that energy $E$ belongs resolvent operator. Starting from results already obtained spectrum in continuous case, we show discrete case generic bounded measurable cocycle has Cantor spectrum.

10.3934/dcdsb.2008.9.541 article EN Discrete and Continuous Dynamical Systems - B 2008-02-01

10.1007/s11554-020-00942-7 article EN Journal of Real-Time Image Processing 2020-01-24

Text proposal has been gaining interest in recent years due to the great success of object categoriesindependent localization. In this paper, we present a novel text-specific technique that provides superior bounding boxes for accurate text localization scenes. The proposed technique, which call Edge Box (TEB), uses binary edge map, gradient map and an orientation image as inputs. Connected components are first found within scored by two low-cue features extracted respectively. These scores...

10.1109/wacv.2017.149 preprint EN 2017-03-01

10.5220/0012402500003660 article EN cc-by-nc-nd Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications 2024-01-01

6G wireless networks are expected to provide seamless and data-based connections that cover space-air-ground underwater networks. As a core partition of future networks, Space-Air-Ground Integrated Networks (SAGIN) have been envisioned countless real-time intelligent applications. To realize this, promoting AI techniques into SAGIN is an inevitable trend. Due the distributed heterogeneous architecture SAGIN, federated learning (FL) then quantum FL emerging model training for enabling...

10.48550/arxiv.2411.01312 preprint EN arXiv (Cornell University) 2024-11-02

This paper proposes a novel intelligent human activity recognition (HAR) framework based on new design of Federated Split Learning (FSL) with Differential Privacy (DP) over edge networks. Our FSL-DP leverages both accelerometer and gyroscope data, achieving significant improvements in HAR accuracy. The evaluation includes detailed comparison between traditional (FL) our FSL framework, showing that the outperforms FL models accuracy loss metrics. Additionally, we examine privacy-performance...

10.48550/arxiv.2411.06263 preprint EN arXiv (Cornell University) 2024-11-09

This paper studies a new latency optimization problem in unmanned aerial vehicles (UAVs)-enabled federated learning (FL) with integrated sensing and communication. In this setup, distributed UAVs participate model training using sensed data collaborate base station (BS) serving as FL aggregator to build global model. The objective is minimize the system over UAV networks by jointly optimizing UAVs' trajectory resource allocation of both BS. formulated troublesome solve due its non-convexity....

10.48550/arxiv.2411.08918 preprint EN arXiv (Cornell University) 2024-11-01

Industrial networks are undergoing rapid transformation driven by the convergence of emerging technologies that revolutionizing conventional workflows, enhancing operational efficiency, and fundamentally redefining industrial landscape across diverse sectors. Amidst this revolution, Digital Twin (DT) emerges as a transformative innovation seamlessly integrates real-world systems with their virtual counterparts, bridging physical digital realms. In article, we present comprehensive survey...

10.48550/arxiv.2412.00209 preprint EN arXiv (Cornell University) 2024-11-29

This study introduces ColonNeXt, a novel fully convolutional attention-based model for polyp segmentation from colonoscopy images, aimed at the enhancing early detection of colorectal cancer. Utilizing purely neural network (CNN), ColonNeXt integrates an encoder-decoder structure with hierarchical multi-scale context-aware (MSCAN) in encoder and block attention module (CBAM) decoder. The decoder further includes proposed CNN-based feature mechanism selective enhancement, ensuring precise...

10.1007/s10278-024-01342-0 article EN Deleted Journal 2024-12-10

This paper presents a state-of-the-art and performance evaluation of real-time text detection methods, having particular focus on the family Laplacian Gaussian (LoG) operators with scale-invariance. The computational complexity is discussed an adaptation to obtained through scale-space representation. In addition, groundtruthing process characterization protocol are proposed, driven repeatability processing time. highlights near-exact approximation at one two orders magnitude execution...

10.5220/0007361503440353 article EN cc-by-nc-nd Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications 2019-01-01

A compact and precise application of rice disease classification is helpful to assist farmers in their work for treatment on the plants therefore could be quick accurate measure eliminate effects diseases more profitably.In past, works were completed by naked-eye observation basically relied experiences.Even so, results are quite subjective heuristic.In this paper, a mobile automatically classify several kinds from plant images then accurately recommend uses pesticides or chemicals.To do...

10.15439/2022r34 article EN cc-by Annals of Computer Science and Information Systems 2022-02-20
Coming Soon ...