A5100422009- Autonomous Vehicle Technology and Safety
- Advanced Neural Network Applications
- Video Surveillance and Tracking Methods
- Traffic Prediction and Management Techniques
- Traffic control and management
- Vehicle emissions and performance
- Robotics and Sensor-Based Localization
- Optical Polarization and Ellipsometry
- Railway Engineering and Dynamics
- Air Quality and Health Impacts
- Vehicle Dynamics and Control Systems
- Traffic and Road Safety
- Catalytic Processes in Materials Science
- Machine Learning in Bioinformatics
- RNA and protein synthesis mechanisms
- Granular flow and fluidized beds
- COVID-19 diagnosis using AI
- High-Voltage Power Transmission Systems
- Hygrothermal properties of building materials
- Text and Document Classification Technologies
- Multimodal Machine Learning Applications
- Dam Engineering and Safety
- Metal Extraction and Bioleaching
- Bauxite Residue and Utilization
- Soil, Finite Element Methods
Nanjing Agricultural University
2025
Taiyuan University of Science and Technology
2023-2024
Fudan University
2024
Jiangsu University
2011-2023
Shandong University of Science and Technology
2023
Ministry of Transport
2022-2023
Research Institute of Highway
2022-2023
Wuhan University of Technology
2011-2022
Tianjin University of Technology and Education
2016-2022
Northwest A&F University
2021
3D Gaussian Splatting (3D-GS) has emerged as a significant advancement in the field of Computer Graphics, offering explicit scene representation and novel view synthesis without reliance on neural networks, such Neural Radiance Fields (NeRF). This technique found diverse applications areas robotics, urban mapping, autonomous navigation, virtual reality/augmented reality, just name few. Given growing popularity expanding research Splatting, this paper presents comprehensive survey relevant...
The binding of transcription factors (TFs) to TF sites plays a vital role in the process regulating gene expression and evolution. With development machine learning deep learning, some successes have been achieved predicting sites. In this paper, we develop model, BTFBS, which predicts whether bacterial combine or not. model takes both amino acid sequences nucleotide as inputs, extracts features through convolutional neural network MultiheadAttention. For use two negative sample sampling...
<p>Artificial intelligence (AI) is driving transformative changes in the field of medicine, with its successful application relying on accurate data and rigorous quality standards. By integrating clinical information, pathology, medical imaging, physiological signals, omics data, AI significantly enhances precision research into disease mechanisms patient prognoses. technologies also demonstrate exceptional potential drug development, surgical automation, brain-computer interface (BCI)...
Concerning roadside traffic detection applications, and to address the millimeter-wave radar's missing data problem caused by target occlusion or absence of features in low-speed conditions, this paper proposes a trajectory compensation method regarding car-following behavior. Referring installation scheme detector, coordinate transformation is presented unify radar spatial coordinates with road coordinates. Considering driver's behavior, optimal velocity model (OV), full difference (FVD),...
Vehicle detection is a crucial task for autonomous driving and demands high accuracy real-time speed. Considering that the current deep learning object model size too large to be deployed on vehicle, this paper introduces lightweight network modify feature extraction layer of YOLOv3 improve remaining convolution structure, improved Lightweight YOLO reduces number parameters quarter. Then, license plate detected calculate actual vehicle width distance between vehicles estimated by width. This...
The red jujube quality is closely associated with its place of origin. In order to quickly and easily identify the geographical origin jujube, classification samples' near-infrared reflectance (NIR) spectra was performed using several fuzzy clustering methods in combination principal component analysis (PCA) linear discriminant (LDA). Firstly, a NIR-M-R2 portable spectrometer used collect four varieties samples from representative producing areas provinces: Gansu, Henan, Shanxi Xinjiang...
Weakly supervised object detection has recently received much attention, since it only requires image-level labels instead of the bounding-box consumed in strongly learning. Nevertheless, save labeling expense is usually at cost model accuracy. In this paper, we propose a simple but effective weakly collaborative learning framework to resolve problem, which trains learner and jointly by enforcing partial feature sharing prediction consistency. For detection, taking WSDDN-like architecture as...
Predicting human trajectory is one of the key knowledge required for autonomous driving and social robots in real scenarios. Recent studies based on Transformer networks have shown a great ability to model behaviors. As far as we know, global information has an essential influence prediction at certain step. However, these methods only rely previous states/attention but ignore important following each pedestrian, which will generally collapse some irregular movements (e.g. acceleration,...
Automobile driver's intention recognition is used to predict what a driver will do the next moment. Generally, driving warning system only considers environment perception, of safe distance and methods, it rarely takes into account. Therefore, its effectiveness accuracy not satisfied. According acceleration characteristic, we analyze operation behavior vehicle status information, propose data sequences about data. The are describe successive present before, model added by can improve...
The transient overvoltage appears commonly in LCC-HVDC infeed power system and could lead to tripping of adjacent PV cascading failures, especially the peak stage. current solutions mainly focus on improving short circuit capacity response reactive compensation devices. However, these need extra devices or not rapidly voltage peak. Thus, this paper proposes an additional emergency control strategy suppress peak, which utilizes PV's rapid adjustment capacity. topology a practical controls key...
Efficient acquisition of real-world embodied data has been increasingly critical. However, large-scale demonstrations captured by remote operation tend to take extremely high costs and fail scale up the size in an efficient manner. Sampling episodes under a simulated environment is promising way for collection while existing simulators high-fidelity modeling on texture physics. To address these limitations, we introduce RoboGSim, real2sim2real robotic simulator, powered 3D Gaussian Splatting...
Large-scale deep neural networks (DNNs), such as large language models (LLMs), have revolutionized the artificial intelligence (AI) field and become increasingly popular. However, training or fine-tuning requires substantial computational power resources, where memory capacity of a single acceleration device like GPU is one most important bottlenecks. Owing to prohibitively overhead (e.g., $10 \times$) GPUs' native allocator, DNN frameworks PyTorch TensorFlow adopt caching allocator that...
Background: The binding of transcription factors (TFs) to TF-binding sites plays a vital role in the process regulating gene expression and evolution. With development machine learning deep learning, some successes have been achieved predicting sites. Then natural question arises: for given factor site, do they bind? This is main motivation this work. Results: In paper, we develop model BTFBS, which predicts whether bacterial combine or not. takes both amino acid sequences nucleotide as...
Both vehicle speed and road geometry are necessary in determining the vehicle's risky behavior. The traditional ACC regulates considering instantaneous effects of conditions, but ignoring geometrical features oncoming accompanied influence on dynamics. So, accidents likely to happen these roads. To elaborate a variable geometry, model is analyzed whether can regulate fitting with ensure ride safety comfort. Hence, we developed an intelligent adaption controller by using Terminal Sliding Mode...