A5100768236- Advanced Image Fusion Techniques
- Remote-Sensing Image Classification
- Image Processing Techniques and Applications
- Natural Language Processing Techniques
- Speech Recognition and Synthesis
- Advanced Image and Video Retrieval Techniques
- Remote Sensing and Land Use
- Data Management and Algorithms
- Photoacoustic and Ultrasonic Imaging
- Speech and Audio Processing
- Image Enhancement Techniques
- Quantum chaos and dynamical systems
- Topic Modeling
- Advanced Differential Equations and Dynamical Systems
- Impact of Technology on Adolescents
- Data Quality and Management
- Caching and Content Delivery
- Mosquito-borne diseases and control
- Infrared Target Detection Methodologies
- Peer-to-Peer Network Technologies
- Advanced Neural Network Applications
- Context-Aware Activity Recognition Systems
- Music and Audio Processing
- Face recognition and analysis
- Opportunistic and Delay-Tolerant Networks
Luoyang Normal University
2014-2024
Xinyang Normal University
2017-2024
North University of China
2024
Jinan University
2023
Chinese Academy of Medical Sciences & Peking Union Medical College
2020-2023
Shandong Normal University
2014-2023
Shantou University
2023
Shantou University Medical College
2023
Nara Institute of Science and Technology
2021-2022
Sun Yat-sen University
2021
Even though the deep neural networks have strong feature representation capability and high recognition accuracy in finger vein recognition, models are computationally intensive poor timeliness. To address these issues, this article proposes a lightweight algorithm for image matching. The proposed uses convolutional model backbone network employs triplet loss function to train model, which not only improves matching accuracy, but also satisfies real-time requirements. In addition,...
Vehicle object detection in aerial scenes has important applications both military and civilian fields. Recently, deep learning shown clear advantages detection, the performance been continuously improved. However, these algorithms rely on anchor-based approaches accompanied by complex convolutional operations. In this paper, we establish a lightweight vehicle algorithm based method of anchor-free. The anchor-free effectively gets rid limitation model capability size fixed anchor box, which...
Based on the theoretical model for "stress-sleep" relationship, this study investigated impact of parental phubbing adolescent sleep quality problems and a moderated mediation mechanism.A total 781 adolescents was surveyed using Chinese version Parental Phubbing Scale, Ultra-brief Screening Scale Depression Anxiety Self-Control Questionnaire children, Pittsburgh Sleep Quality Index Scale.Parental negative emotions were significantly positively correlated to problems, but self-control not...
Achieving Artificial General Intelligence (AGI) requires AI agents that can not only make stratigic decisions but also engage in flexible and meaningful communication. Inspired by Wittgenstein's language game theory Philosophical Investigations, we propose learn through in-context interaction rather than traditional multi-stage frameworks separate decision-making from expression. Using Werewolf, a social deduction tests understanding, strategic interaction, adaptability, develop the...
Manual inspection of histopathological images is important in clinical cancer diagnosis. Pathologists implement pathological diagnosis and prognostic evaluation through the microscopic examination slices. This entire process time-consuming, laborious, challenging for pathologists. The modern use whole-slide imaging, which scans histopathology slides to digital slices, analysis using computer-aided an essential problem.To solve problem difficult labeling data, improve flexibility...
Mounting evidence indicates that Zika virus (ZIKV) is closely related to neurological disorders such as microcephaly and Guillain-Barré syndrome. There are currently no effective vaccines FDA-approved inhibitors against ZIKV infection. The flaviviral heterodimeric serine protease NS2B-NS3 plays an essential role in maturation replication, thus becoming a promising target anti-ZIKV therapy. Herein, we developed fluorescence-based screening assay search for targeting the (ZIKVpro), identified...
Instruments detect and respond to various parameters, playing a crucial role in civil industrial applications. However, the current instrument recognition algorithm still suffers from low detection accuracy, large amount of model calculation, degree automatic recognition. In this paper, we propose an efficient real-time indication acquisition based on lightweight deep convolutional neural network hybrid attention fine-grained features. algorithm, divide identification into three sub-tasks,...
All RNA viruses encode the RNA-dependent polymerase (RdRp) which replicates and transcribes viral RNA. This essential enzyme does not exist in mammalian cells, thus presents a main target for development of antiviral drugs with potential pan-antiviral activity. In this study, we take advantage structurally equivalent site dengue virus (DENV) RdRp, N-pocket, human norovirus (hNV) B-site, performed parallel structure-based virtual screening to discover compounds that can inhibit RdRps both hNV...
Learning-based heuristics for solving combinatorial optimization problems has recently attracted much academic attention. While most of the existing works only consider single objective problem with simple constraints, many real-world have multiobjective perspective and contain a rich set constraints. This paper proposes deep reinforcement learning evolutionary algorithm typical complex called vehicle routing time windows (MO-VRPTW). In proposed algorithm, decomposition strategy is applied...
Vehicle detection in aerial photography scenarios has a wide range of promising applications the military and civilian fields. Recently, object algorithms based on depth models have shown superior performance vehicle tasks. However, these are often accompanied by large amount computation resource consumption, which leads to inability perform real-time detection. In addition, insufficient feature extraction capability complex background information also lead low accuracy. this letter, we...
SUMMARY In this paper, we mainly focus on similarity joins massive high‐dimensional vectors, it is a costly operation because of curse dimensionality. The main idea our proposed solution to design novel filtering method that can filter out as many vector pairs possible at relative low cost. By using the good dimension reduction ability piecewise aggregate approximation and symbolic techniques, three approaches deal with high dimensional‐vector join query: single‐PAA–based query,...
Abstract Background The major public health crisis caused by the rapid spread of coronavirus disease 2019 (COVID-19) and large-scale measures such as social isolation school closures enforced some countries have severely affected on physical mental wellbeing children adolescents globally. This study aimed to estimate prevalence psychological impact investigate similarities differences in influential factors for generalized anxiety disorder among rural a relatively lesser noticed population...