A5101525745- Music and Audio Processing
- Speech and Audio Processing
- Electric Power System Optimization
- Simulation Techniques and Applications
- Visual perception and processing mechanisms
- Seismic Imaging and Inversion Techniques
- Anomaly Detection Techniques and Applications
- Retinal Development and Disorders
- Iterative Learning Control Systems
- Image Enhancement Techniques
- Water resources management and optimization
- Human Pose and Action Recognition
- Advanced Multi-Objective Optimization Algorithms
- Video Surveillance and Tracking Methods
- Drilling and Well Engineering
- Stability and Control of Uncertain Systems
- Music Technology and Sound Studies
- Radiomics and Machine Learning in Medical Imaging
- Advanced Algorithms and Applications
- Advanced Image Processing Techniques
- Cerebrovascular and Carotid Artery Diseases
- Image and Video Stabilization
- Advanced Numerical Analysis Techniques
- Face recognition and analysis
- Speech Recognition and Synthesis
South China University of Technology
2012-2024
Shanghai Jiao Tong University
2020-2024
Sun Yat-sen Memorial Hospital
2022-2024
Sun Yat-sen University
2022-2024
Lanzhou Jiaotong University
2024
Beijing University of Chinese Medicine
2023
Sichuan University
2023
West China Hospital of Sichuan University
2023
Hohai University
2023
Shenzhen Pingle Orthopedic Hospital
2023
To develop a general unsupervised anomaly detection method based only on MR images of normal brains to automatically detect various brain abnormalities. In this study, novel three-dimensional deep autoencoder network is proposed and segment abnormalities without being trained any abnormal samples. A total 578 T2w volumes obvious were used for model training validation. The 3D was evaluated two different datasets (BraTs dataset in-house dataset) containing from patients with glioblastoma,...
Drone delivery is an emerging service at the early adoption stage. It imperative to understand what it takes for public accept such services. This study asks mechanism influences people's switching intention in e-commerce drone A 7 points Likert scale questionnaire were developed after two rounds of pretest and a total 83 surveys collected from business school USA. applies privacy calculus theory technology anxiety innovation develop research model. An empirical survey structural equation...
In aerospace engineering, condition monitoring is an important reference for evaluating the performance of complex systems. Especially, effective anomaly detection, based on telemetry data, plays role system health management a spacecraft. With advantages easy-to-use, high efficiency, and data-driven, predicted model has been applied anomalous point detection data. However, compared with abnormal mode, fragment more attractive meaningful identification. Therefore, strategy proposed...
Deep learning based approaches have greatly improved the performance of spoken keyword spotting (KWS). However, KWS different languages should their own corresponding modeling units to optimize performance. In this paper, we propose an end-to-end Mandarin system using Convolutional Recurrent Neural Network with Connectionist Temporal Classification (CTC) loss function (CRNN-CTC). The tonal syllables are adopted as units. Experiments on AISHELL-2 datasets showed that proposed approach tasks...
Pavement icing imposes a great threat to driving safety and impacts the efficiency of road transportation system in cold regions. This has attracted research predicting pavement time solve problems brought about by icing. Different models have been proposed past decades predict icing, within which support vector regression (SVR) is widely used algorithm for calibrating highly nonlinear relationships. paper presents hybrid improved SVR with an enhancement operation response surface method...
We consider the problem of producing confidence intervals for mean response a system represented by stochastic simulation that is driven input models have been estimated from "real-world" data. Therefore, we want interval to account both uncertainty about and noise in output; standard practice only accounts noise. To achieve this goal introduce metamodel-assisted bootstrapping, illustrate its performance relative other proposals dealing with on two queueing examples.
Tanzania is located in eastern Africa with a predominantly agricultural ecomomy, the potential for developing and utilizing cultivated land are promising, but scientific guidance required. B, Zn Se essential micronutrients plants human body crucial biological functions, particular, significant health considered as "the king of anti-cancer". As these elements required by or mainly absorbed from soil directly indirectly, therefore, it important to understand contents distributions them guiding...
Paddy field algorithm (PFA) is a fast random with global search capability. But when the number of solutions over range, efficiency becomes low because it executes lot redundant iterations. Pattern sensitive to initial condition. In order strength local ability, paddy introduced and novel proposed in this paper. The hybrid operates by initially scattering seeds at parameter space. final result found pattern based on PFA algorithm. This tested three sample functions alongside basic method....
Direct yaw moment control (DYC) and differential drive-assist steering (DDAS) for distributed-drive vehicles are both realized by allocating the in-wheel motor torque. To address interference caused overlapping objectives, this paper proposes a multilayer strategy that integrates DYC DDAS, consisting of an upper controller, coordinated decision layer, torque distribution layer. The designed based on vehicle’s dynamic characteristics, incorporates adaptive fuzzy system dual PID DDAS system....
Power generation dispatching is a large complex system problem with multi-dimensional and nonlinear characteristics. A mathematical model was established based on the principle of reservoir operation. quantity optimal scheduling processes were obtained by calculating daily runoff process within three typical years, number simulated using progressive optimality algorithm (POA) in combination genetic (GA). After analyzing processes, corresponding rules determined, practical formulas obtained....
Violence detection in videos can help maintain public order, detect crimes, or provide timely assistance. In this paper, we aim to leverage multimodal information determine whether successive frames contain violence. Specifically, propose an audiovisual dependency attention (AVD-attention) module modified from the co-attention architecture fuse visual and audio information, unlike commonly used methods such as feature concatenation, addition, score fusion. Because AVD-attention module's map...
This paper presents a navigation algorithm based on interval type-2 fuzzy neural network fitting Q-learning (IT2FNN-Q), and succeeds in providing solution for mobile robot complex environments. The utilizes the reasoning adaptive ability extensive functional app roximation features of IT2FNN to solve this problem, mapped from state space action space, unknown Compared with BP (BP-Q), IT2FNN-Q endows better real-time decision-making abilities solves slow convergence nonconvergence problems,...
Gait is influenced by race, age, and diseases type. Reference values for gait are closely related to numerous health outcomes. To gain a comprehensive understanding of patterns, particularly in relation race-related pathologies disorders, it crucial establish reference daily life considering sex age. Therefore, our objective was present age-based life, providing valuable foundation further research clinical applications.To lower extremity joint kinematics kinetics during asymptomatic adult...
Increasingly, deformation prediction has become an essential research topic in sluice safety control, which requires significant attention. However, there is still a lack of practical and efficient modeling for deformation. In order to address the limitations mining deep features long-time data series traditional statistical model, this paper, improved long short-term memory (LSTM) model weighted Markov are introduced predict method, seagull optimization algorithm (SOA) utilized optimize...
To investigate the relative sensory and perceptual contributions to central visual function of patients with retinitis pigmentosa (RP), we tested symmetry discrimination using block patterns varying types symmetric organization. Eleven control subjects normal vision 11 RP 20/30 acuity or better, viewed presented for 255 msec. The differed in type organization were required identify type. performed significantly better (89.2%) than (74.5%). Four hypotheses account these findings results as...
S-shape acceleration/deceleration algorithms play a key role in the development of high speed CNC.This paper first describes basic formulation Acc/Dec, which jerk, acceleration, velocity, displacement are discussed detail. Based on above considerations, novel Acc/Dec algorithm equipped with rounding error compensation tactic is employed compensates at transition each segment. Moreover proposed can reduce quantization effectively and look-ahead technique also mentioned this section....