A5087731729- Thermography and Photoacoustic Techniques
- Human-Automation Interaction and Safety
- EEG and Brain-Computer Interfaces
- Autonomous Vehicle Technology and Safety
- Functional Brain Connectivity Studies
- Industrial Vision Systems and Defect Detection
- Neural dynamics and brain function
- Cellular Automata and Applications
- Advanced Image Processing Techniques
- Neural Networks and Applications
- Topological Materials and Phenomena
- Image and Signal Denoising Methods
- Reliability and Maintenance Optimization
- Retinal Imaging and Analysis
- Nonlinear Dynamics and Pattern Formation
- Spectroscopy and Chemometric Analyses
- Ultrasonics and Acoustic Wave Propagation
- Video Surveillance and Tracking Methods
- Sleep and Work-Related Fatigue
- Traffic and Road Safety
- Fault Detection and Control Systems
- Occupational Health and Safety Research
- Probabilistic and Robust Engineering Design
- Remote Sensing in Agriculture
- Machine Fault Diagnosis Techniques
Cranfield University
2016-2025
Qingdao Agricultural University
2023-2025
Northwestern Polytechnical University
2024-2025
Centre for Life
2021-2025
Xi'an University of Science and Technology
2021-2025
Ningbo University
2024-2025
Harbin Institute of Technology
2020-2025
Hong Kong Polytechnic University
2025
Hebei University of Engineering
2025
Sericultural Research Institute
2025
The existing deep learning-based Personal Protective Equipment (PPE) detectors can only detect limited types of PPE and their performance needs to be improved, particularly for deployment on real construction sites. This paper introduces an approach train evaluate eight learning detectors, application purposes, based You Only Look Once (YOLO) architectures six classes, including helmets with four colours, person, vest. Meanwhile, a dedicated high-quality dataset, CHV, consisting 1330 images,...
It is important to conduct grassland resource surveys for the scientific management of resources. Currently, remote sensing technology widely used classify land cover. The fine classification datasets grasslands with high spatial and temporal resolutions are very necessary research. In order use data conveniently, this study selected Google Earth Engine platform select 100-m resolution PROBA-V images from 2018 Zambia, in central Africa. differences normalized vegetation index time-series...
In present-day highly-automated vehicles, there are occasions when the driving system disengages and human driver is required to take-over. This of great importance a vehicle's safety ride comfort. U.S state California, Autonomous Vehicle Testing Regulations require every manufacturer testing autonomous vehicles on public roads submit an annual report summarizing disengagements technology experienced during testing. On 1 January 2016, seven manufacturers submitted their first disengagement...
Alzheimer's disease (AD) accounts for 60%-70% of all dementia cases, and clinical diagnosis at its early stage is extremely difficult. As several new drugs aiming to modify progression or alleviate symptoms are being developed, assess their efficacy, novel robust biomarkers brain function urgently required. This paper aims explore a routine gain such using the quantitative analysis electroencephalography (QEEG). proposes supervised classification framework that uses EEG signals classify...
Automated detection of vascular structures is great importance in understanding the mechanism, diagnosis, and treatment many pathologies. However, automatic continues to be an open issue because difficulties posed by multiple factors, such as poor contrast, inhomogeneous backgrounds, anatomical variations, presence noise during image acquisition. In this paper, we propose a novel 2-D/3-D symmetry filter tackle these challenging issues for enhancing vessels from different imaging modalities....
In this paper, we propose a talking face generation method that takes an audio signal as input and short target video clip reference, synthesizes photo-realistic of the with natural lip motions, head poses, eye blinks are in-sync signal. We note synthetic attributes include not only explicit ones such motions have high correlations speech, but also implicit poses weak correlation audio. To model complicated relationships among different audio, FACe Implicit Attribute Learning Generative...
The simultaneous recognition of multiple objects in one image remains a challenging task, spanning events the field such as various object scales, inconsistent appearances, and confused inter-class relationships. Recent research efforts mainly resort to statistic label co-occurrences linguistic word embedding enhance unclear semantics. Different from these researches, this paper, we propose novel Transformer-based Dual Relation learning framework, constructing complementary relationships by...
The urban surface temperature is a complex integrated natural-human geographic phenomena; with the development of geostatistical methods and application multisource data, its research has gradually shifted from single perspective to study that integrates multiple factors such as nature humanity. However, based on context integration natural human mutual constraints each factor, mechanism influence habitat thermal environment needs be further deepened. Therefore, this paper explores spatial...
Accurate dynamic driver head pose tracking is of great importance for driver–automotive collaboration, intelligent copilot, head-up display (HUD), and other human-centered automated driving applications. To further advance this technology, article proposes a low-cost markerless head-tracking system using deep learning-based estimation model. The proposed requires only red, green, blue (RGB) camera without hardware or markers. enhance the accuracy driver’s estimation, spatiotemporal vision...
Few-shot class-incremental learning (FSCIL) aims at to classify new classes continually from limited samples without forgetting the old classes. The mainstream framework tackling FSCIL is first adopt cross-entropy (CE) loss for training base session, then freeze feature extractor adapt However, in this work, we find that CE not ideal session as it suffers poor class separation terms of representations, which further degrades generalization novel One tempting method mitigate problem apply an...
In quantum anomalous Hall (QAH) insulators, the interior is insulating but electrons can travel with zero resistance along one-dimensional (1D) conducting paths known as chiral edge channels (CECs). These CECs have been predicted to be confined 1D edges and exponentially decay in two-dimensional (2D) bulk. this Letter, we present results of a systematic study QAH devices fashioned bar geometry different widths under gate voltages. At charge neutral point, effect persists device width only...
Background Among patients who are discharged from the Emergency Department (ED), about 3% return within 30 days. Revisits can be related to nature of disease, medical errors, and/or inadequate diagnoses and treatment during their initial ED visit. Identification high-risk patient population help device new strategies for improved care with reduced utilization. Methods Findings A decision tree based model discriminant Electronic Medical Record (EMR) features was developed validated,...
Iceberg calving is a major component of the total mass balance Greenland ice sheet (GrIS). A century-long record icebergs comes from International Ice Patrol's (I48N) passing latitude 48° N, off Newfoundland. I48N exhibits strong interannual variability, with significant increase in amplitude over recent decades. In this study, we show, through combination nonlinear system identification and coupled ocean-iceberg modelling, that I48N's variability predominantly caused by fluctuation GrIS...
Leakage in retinal angiography currently is a key feature for confirming the activities of lesions management wide range diseases, such as diabetic maculopathy and paediatric malarial retinopathy. This paper proposes new saliency-based method detection leakage fluorescein angiography. A superpixel approach firstly employed to divide image into meaningful patches (or superpixels) at different levels. Two saliency cues, intensity compactness, are then proposed estimation map each individual...
Automatic and flexible geometry measurement of the weld pool surface can help better understand complex welding processes even provide feedback to control this process. Most existing imaging systems use an additional source illumination remove light interference coming from arc but it is usually costly. This paper introduces a novel low-cost optical-sensor-based monitoring system working under passive mode monitor wire + additive manufacture process, particularly for plasma welding....