- Video Surveillance and Tracking Methods
- Chaos-based Image/Signal Encryption
- Advanced Steganography and Watermarking Techniques
- Optical Network Technologies
- Traffic Prediction and Management Techniques
- Autonomous Vehicle Technology and Safety
- Medical Image Segmentation Techniques
- Photonic and Optical Devices
- Parallel Computing and Optimization Techniques
- Energy Harvesting in Wireless Networks
- Functional Brain Connectivity Studies
- Image and Signal Denoising Methods
- Energy Efficient Wireless Sensor Networks
- Image Enhancement Techniques
- Embedded Systems Design Techniques
- Advanced Image and Video Retrieval Techniques
- Neural Networks and Applications
- Advanced MRI Techniques and Applications
- RFID technology advancements
- Digital Media Forensic Detection
- Blind Source Separation Techniques
- Image Retrieval and Classification Techniques
- Interconnection Networks and Systems
- User Authentication and Security Systems
- Antenna Design and Analysis
Nanyang Technological University
2014-2024
Nanyang Polytechnic
2008-2023
Iran University of Science and Technology
1995
A non-contact, non-invasive monitoring system to measure and estimate the heart breathing rate of humans using a frequency-modulated continuous wave (FMCW) mm-wave radar at 77 GHz is presented. novel diagnostic proposed which extracts heartbeat phase signals from FMCW (reconstructed Fourier series analysis) test three-layer artificial neural network model predict presence arrhythmia in individuals. The effect person orientation, distance measurement movement was analyzed with respect...
In this paper, machine learning models for an effective estimation of soil moisture content using a microwave short-range and wideband radar sensor are proposed. The is measured as the volumetric water off-the-shelf operating at 3–10 GHz. captures reflected signals that post processed to determine which mapped input features extracted from training models. addition, results compared analyzed with contact-based Vernier sensor. Different trained neural network, support vector machine, linear...
Traffic information is an important tool in the planning, maintenance, and control of any modern transport system. Of special interest to traffic engineers are parameters flow such as volume, speed, type vehicle, queue parameters, movements at junctions, etc. Various algorithms, mainly based on background differencing techniques, have been applied for this purpose. Since background-based algorithms very sensitive ambient lighting conditions, they not yielded expectative results. In paper, we...
The real-time measurement of various traffic parameters including queue is required in many situations such as accident and congestion monitoring adjusting the timings lights. In case detection, at least two algorithms have been proposed by previous researchers. Those are used for detection unable to measure parameters. authors propose a method based on applying combination noise insensitive simple number sub-profiles (a one-pixel-wide key-region) along road. algorithm consists motion...
Personal identification numbers, passwords, smart cards and digital certificates are some of the means employed for user authentication in various electronic commerce applications. However, these do not really identify a person, but only knowledge data or belonging determined object. This paper introduces notion biometric signature – new approach to integrate biometrics with public key infrastructure, using based generation which is secure, efficacious, fast, convenient, non‐invasive...
Developing an effective and efficient electroencephalography (EEG)-based drowsiness monitoring system is crucial for enhancing road safety reducing the risk of accidents. For general usage, cross-subject evaluation indispensable. Despite progress in unsupervised domain adaptation (UDA) source-free (SFDA) methods, these often rely on availability labeled source data or white-box models, posing potential privacy risks. This study explores a more challenging setting UDA EEG-based detection,...
In this paper, a perceptually scalable and JPEG compression tolerant DCT-based image encryption scheme is proposed. There are two main features of the proposed scheme, first decryption algorithm able to reconstruct plaintext-image even if encrypted compressed. Secondly, can be adjusted produce cipher-image with varying perceptual distortion. The relies on some very interesting properties orthogonal matrices containing columns that form set orthonormal basis vectors. It envisaged cryptosystem...
In recent years, the development of smart transportation has accelerated research on semantic segmentation as it is one most important problems in this area. A large receptive field always been center focus when designing convolutional neural networks for segmentation. majority techniques have used maxpooling to increase a network at an expense decreasing its spatial resolution. Although idea shown improved results object detection applications, however, comes segmentation, high resolution...
For the past few years, a steady growth has been observed in Internet-based commerce activities Singapore and Asia-Pacific region. However, there very limited study that empirically investigates socio-economic factors their influence on adoption of electronic (e-commerce) by consumers Singapore. This seeks to fill some gaps this area. Viewing such medium as form new innovation, five characters namely gender, age, income level, education level exposure Internet were hypothesized see whether...
A novel low-profile tri-band bandpass frequency-selective surface (FSS) is proposed in this paper. The structure designed by the two-dimensional repetition of a unit cell consisting four apertures. Each aperture combination T-shaped, U-shaped and rectangular-shaped aligned horizontally or vertically. performance FSS independent angle incidence incident wave due to orientation apertures cell. operates at desired frequency bands close 8-14 GHz. insertion loss 3.26, 2.68 1.12 dB respective...
In this paper we present a framework for enhancing trust in Internet commerce. Experience shows that efficient cryptographic protocols are not enough to guarantee peoples’ confidence commerce; the transacting parties must also each other. Hence, main ingredient missing today’s e‐commerce infrastructures is modeling and implementing trust. Several attempts have been made provide secure trusted but few seen any practical use. This how can be provided through network of Trust Service Providers...