A5057248354- Video Surveillance and Tracking Methods
- Advanced Optical Network Technologies
- Advanced Neural Network Applications
- Advanced Photonic Communication Systems
- Optical Network Technologies
- Semantic Web and Ontologies
- Autonomous Vehicle Technology and Safety
- Image Retrieval and Classification Techniques
- Advanced Computational Techniques and Applications
- Advanced Image and Video Retrieval Techniques
- Service-Oriented Architecture and Web Services
- Advanced Database Systems and Queries
- Software-Defined Networks and 5G
- Image Processing and 3D Reconstruction
- Advanced Image Fusion Techniques
- Vehicular Ad Hoc Networks (VANETs)
- Face and Expression Recognition
- Simulation and Modeling Applications
- Advanced Vision and Imaging
- Anomaly Detection Techniques and Applications
- Image and Object Detection Techniques
- Advanced Measurement and Detection Methods
- Human Pose and Action Recognition
- Robotics and Sensor-Based Localization
- Vehicle License Plate Recognition
Chuzhou University
2025
Anhui University
2025
Beijing Union University
2015-2024
East China Normal University
2023-2024
New York University Shanghai
2023-2024
Xidian University
2023
Guilin University of Electronic Technology
2022
Ministry of Industry and Information Technology
2022
Chinese Academy of Geological Sciences
2022
Pudong Medical Center
2022
Nirmatrelvir-ritonavir has been authorized for emergency use by many countries the treatment of coronavirus disease 2019 (Covid-19). However, supply falls short global demand, which creates a need more options. VV116 is an oral antiviral agent with potent activity against severe acute respiratory syndrome 2 (SARS-CoV-2).We conducted phase 3, noninferiority, observer-blinded, randomized trial during outbreak caused B.1.1.529 (omicron) variant SARS-CoV-2. Symptomatic adults mild-to-moderate...
Traffic sign detection is an important component of autonomous vehicles. There still a mismatch problem between the existing algorithm and its practical application in real traffic scenes, which mainly due to accuracy data acquisition. To tackle this problem, study proposed improved sparse R-CNN that integrates coordinate attention block with ResNeSt builds feature pyramid modify backbone, enables extracted features focus on information, improves accuracy. In order obtain more diverse data,...
Ensemble empirical-mode decomposition (EEMD) is a novel adaptive time-frequency analysis method, which particularly suitable for extracting useful information from noisy nonlinear or nonstationary data. Unfortunately, since the EEMD highly compute-intensive, method does not apply in real-time applications on top of commercial-off-the-shelf computers. Aiming at this problem, parallelized has been developed using general-purpose computing graphics processing unit (GPGPU), namely, G-EEMD. A...
Object detection plays a vital role in autonomous driving systems, and the accurate of surrounding objects can ensure safe vehicles. This paper proposes category-assisted transformer object detector called DetectFormer for driving. The proposed achieve better accuracy compared with baseline. Specifically, ClassDecoder is assisted by proposal categories global information from Global Extract Encoder (GEE) to improve category sensitivity performance. fits distribution specific scene...
Abstract During the past decade, cognitive neuroscience has been calling for population diversity to address challenge of validity and generalizability, ushering in a new era neuroscience. The developing Chinese Color Nest Project (devCCNP, 2013–2022), first ten-year stage lifespan CCNP (2013–2032), is two-stages project focusing on brain-mind development. aims create share large-scale, longitudinal multimodal dataset typically children adolescents (ages 6.0–17.9 at enrolment) population....
Small-object detection is a challenging task in computer vision due to the limited training samples and low-quality images. Transfer learning, which transfers knowledge learned from large dataset small dataset, popular method for improving performance on data. However, we empirically find that discrepancy, directly transferring model trained general object small-object datasets obtains inferior performance. In this paper, propose TranSDet, novel approach effective transfer learning...
Recently the research of vehicle detection is mainly through machine learning, but it still has low accuracy problem. With study researchers, using deep learning methods becomes hot. In this paper, a selective search method and target model based on Fast R-CNN are used to detect vehicle. The strategy optimizes by preprocessing sample image new network structure. Firstly, experiment uses public KITTI data set self-collected BUU-T2Y set, respectively, for training validation test. Secondly,...
Vision-based object detection is an essential component of autonomous driving. Because vehicles typically have limited on-board computing resources, a small-sized model required. Simultaneously, high accuracy and real-time inference speeds are required to ensure safety while In this paper, anchor-free lightweight detector for driving called ALODAD proposed. incorporates attention scheme into the neural network GhostNet builds framework achieve lower computational costs provide parameters...
Today, multi-sensor fusion detection frameworks in autonomous driving, especially sequence-based data-level frameworks, face high latency and coupling issues generally perform worse than LiDAR-only detectors. On this basis, we propose PMPF, point-cloud multiple-pixel fusion, for 3D object detection. PMPF projects the point cloud data onto image plane, where region pixels are processed to correspond with points decorated data, such that fused can be applied detectors autoencoders. is a...
ABSTRACT The adaptive inverse distance weighted interpolation technique significantly enhances the calculation accuracy of geological boreholes by employing membership functions to adaptively compute weights. However, these typically utilize conventional mathematical forms, and identifying more suitable based on sample distribution pattern remains a critical research area for improving computational accuracy. This study introduces geostatistical method examine spatial characteristics...
Vehicular networks play an important role in the intelligent transportation systems which have gained technical supports from car industry. Due to mobility and broadcast nature of wireless communication, security vehicular is a critical issue for academia Many solutions been proposed target provisioning. However, most them various shortcomings. Based on elliptic curve public key cryptography algorithm, this paper, we propose new anonymous roaming authentication protocol Long Term...
Augmented Reality (AR) is crucial for immersive Human–Computer Interaction (HCI) and the vision of Artificial Intelligence (AI). Labeled data drives object recognition in AR. However, manually annotating expensive, labor-intensive, distribution asymmetry. Scantily labeled limits application Aiming at solving problem insufficient asymmetry training AR recognition, an automated synthesis method, i.e., background augmentation generative adversarial networks (BAGANs), proposed this paper based...
In complex mining environments, driverless trucks are required to cooperate with multiple intelligent systems. They must perform obstacle avoidance based on factors such as the site road width, type, vehicle body movement state, and ground concavity-convexity. Targeting open-pit area, this paper proposes an object detection (IMOD) model developed using a 5G-multi-UAV deep learning approach. The IMOD employs data sensors monitor surface in real time within multisystem collaborative 5G...
Color constancy is an important perceptual ability of humans to recover the color objects invariant light information. It also necessary for a robust machine vision system. Until now, number algorithms have been proposed in literature. In particular, edge-based uses edge image estimate color. shown be rich framework that can represent many existing illumination estimation solutions with various parameter settings. However, ill-posed problem; every algorithm always given out under some...
Among numerous Chinese keyword extraction methods, characteristics were shortly considered. This phenomenon going against the precision enhancement of extraction. An extended term frequency based method(Extended TF) is proposed in this paper which combined linguistic with basic TF method. Unary, binary and ternary grammars for candidate as well other features all taken into account. The method establishes classification model using support vector machine. Tests show that improved key words...
Industrial nameplates serve as a means of conveying critical information and parameters. In this work, we propose novel approach for rectifying industrial nameplate pictures utilizing Probabilistic Hough Transform. Our method effectively corrects distortions clipping, features collection challenging analysis. To determine the corners nameplate, employ progressive Probability Transform, which not only enhances detection accuracy but also possesses ability to handle complex scenarios. The...
With the continuous breakthrough of object detection, there are many achievements in combination field computer science and mechanical automation. Especially manufacturing industry, detection has played a great role weld inspection link during production vehicle wheels. The algorithm is used to align on wheel hub with air sensor by identifying welds rotate order achieve purpose automatic inspection. application this technology can greatly Improve efficiency accuracy, getting rid...
As we know, feature extraction has an important role in crowd density estimation. In our paper, introduce a new texture called Tamura, which is usually used image retrieval algorithms. On the other hand, time consuming another issue that must be considered, especially for real-time application of most methods, multiple features with high dimension such as gray level co-occurrence matrix (GLCM) are to construct input vector, will decrease performance whole method. order solve problem, use...