- EEG and Brain-Computer Interfaces
- Neuroscience and Neural Engineering
- Advanced Memory and Neural Computing
- Neural dynamics and brain function
- Gaze Tracking and Assistive Technology
- Blind Source Separation Techniques
- Video Coding and Compression Technologies
- Advanced Vision and Imaging
- Neural Networks and Applications
- Advanced Steganography and Watermarking Techniques
- Biometric Identification and Security
- CCD and CMOS Imaging Sensors
- Advanced Data Compression Techniques
- Image and Video Stabilization
- Advanced Image Processing Techniques
- Robotics and Automated Systems
- Image Enhancement Techniques
- Chaos control and synchronization
- Digital Media Forensic Detection
- Embedded Systems Design Techniques
- Face recognition and analysis
- Interconnection Networks and Systems
- Advanced Neural Network Applications
- Adversarial Robustness in Machine Learning
- User Authentication and Security Systems
Tianjin University
2017-2023
Nanjing University of Information Science and Technology
2022
Electronics and Telecommunications Research Institute
2005-2019
Northwestern Polytechnical University
2010
Objective: Recently, electroencephalography (EEG)- based brain-computer interfaces (BCIs) have made tremendous progress in increasing communication speed. However, current BCI systems could only implement a small number of command codes, which hampers their applicability. Methods: This study developed high-speed hybrid system containing as many 108 instructions, were encoded by concurrent P300 and steady-state visual evoked potential (SSVEP) features decoded an ensemble task-related...
. Brain-computer interfaces (BCIs) have recently made significant strides in expanding their instruction set, which has attracted wide attention from researchers. The number of targets and commands is a key indicator how well BCIs can decode the brain's intentions. No studies reported BCI system with over 200 targets.
Abstract Objective. P300s are one of the most studied event-related potentials (ERPs), which have been widely used for brain–computer interfaces (BCIs). Thus, fast and accurate recognition is an important issue BCI study. Recently, there emerges a lot novel classification algorithms P300-speller. Among them, discriminative canonical pattern matching (DCPM) has proven to work effectively, in spatial (DSP) filter can significantly enhance features P300s. However, ERPs space varies with time,...
The brain-computer interface (BCI) spellers based on steady-state visual evoked potentials (SSVEPs) have recently been widely investigated for their high information transfer rates (ITRs). This paper aims to improve the practicability of SSVEP-BCIs high-speed spelling. system acquired electroencephalogram (EEG) data from a self-developed dedicated EEG device and stimulation was arranged as keyboard. task-related component analysis (TRCA) spatial filter modified (mTRCA) target classification...
Brain–computer interfaces (BCI) have witnessed a rapid development in recent years. However, the active BCI paradigm is still underdeveloped with lack of variety. It imperative to adapt more voluntary mental activities for control, which can induce separable electroencephalography (EEG) features. This study aims demonstrate brain function timing prediction, i.e., expectation upcoming time intervals, accessible BCIs. Eighteen subjects were selected this study. They trained precise idea two...
Visual brain-computer interface (BCI) systems have made tremendous process in recent years. It has been demonstrated to perform well spelling words. However, different from English words one-dimension sequences, Chinese characters are often written a two-dimensional structure. Previous studies had never investigated how use BCI 'write' but not 'spell' characters. This study developed an innovative BCI-controlled robot for writing The system contained 108 commands displayed 9*12 array. A...
An application specific processor for an H.264 decoder with a configurable embedded is designed in this research. The motion compensation, inverse integer transform, quantization, and entropy decoding algorithm of software are optimized. We improved the performance instruction-level hardware optimization, which tailored to architecture. optimized instructions video processing can be used other compression standards such as MPEG 1, 2, 4. A significant improvement achieved high flexibility....
An implementation of Yolo-v2 image recognition and other testbenches for a deep learning accelerator is presented. This chip the initial version our on-going effort higher performance development. The based on systolic array can handle convolution max-pooling layer in combined way or separately using 16 bit floating-point data. It also supports inner-product LSTM layers. For demonstration as one design verification testbenches, we implemented 80 object classes. We converted software to...
In this paper, we present a performance analysis for an MPEG-4 video codec based on the on-chip network communication architecture. The existing buses of system-on-a-chip (SoC) have some limitation data traffic bandwidth since large number silicon IPs share bus. An is introduced to solve problem buses, in which concept computer applied architecture SoC. We compared and Advanced Micro-controller Bus Architecture (AMBA) Experimental results show that improved over 50% design multi-layer AMBA
Visual brain-computer interfaces (BCIs) have achieved great progress in speed recently. But the problem of visual fatigue caused by intense flashes poses a challenge designing practical systems for long-term use. A direct way to improve comfort is reduce stimulus contrast. it could also weaken featured evoked potentials, which would bring negative impact on system accuracy. Thus it's significant figure out optimal contrast that both high and This study investigated effects different...
Steady-State Visual Evoked Potentials (SSVEPs) have become one of the most used neural signals for brain- computer interfaces (BCIs) due to their stability and high signal- to-noise rate. However, performance SSVEP-based BCIs would degrade with a few training samples. This study was proposed enhance detection SSVEP by combining supervised learning information from samples unsupervised trial be tested. A new method, i.e. cyclic shift trials (CST), generate calibration test data, which were...
User identity recognition is the key shield to protect users’ privacy data from disclosure and embezzlement. The user of mobile devices such as phones mainly includes fingerprint recognition, nine-grid password, face digital etc. Due requirements computing resources convenience devices, these verification methods have their own shortcomings. In this paper, a technology based on finger trajectory proposed. Based analysis data, feature user's movement extracted realize identification user....
With the popularity of smartphones, it is often easy to maliciously leak important information by taking pictures phone. Robust watermarking that can resist screen photography achieve protection information. Since photo process cause some irreversible distortion, currently available watermarks do not consider image content well and visual quality very high. Therefore, this paper proposes a new screen-photography robust watermark. In terms embedding region selection, intensity-based...
In recent years, visual-based Brain-Computer Interface (BCI) systems have gained significant attention due to their high Information Transfer Rate (ITR). practical applications, there is a growing demand for large instruction set BCI support more complex commands. However, users may experience fatigue-related issues during prolonged engagement in visual tasks, which negatively impacts the modeling accuracy of systems. To address issue signal degradation caused by user subjective intentions...