A5047749179- Atmospheric chemistry and aerosols
- Atmospheric Ozone and Climate
- Air Quality and Health Impacts
- Catalytic Processes in Materials Science
- Infrared Target Detection Methodologies
- Radiation Detection and Scintillator Technologies
- Brain Tumor Detection and Classification
- Protist diversity and phylogeny
- Metabolomics and Mass Spectrometry Studies
- Catalysis and Oxidation Reactions
- Transportation and Mobility Innovations
- Multimodal Machine Learning Applications
- Atmospheric aerosols and clouds
- Advanced Proteomics Techniques and Applications
- Advanced Measurement and Detection Methods
- Sharing Economy and Platforms
- Air Quality Monitoring and Forecasting
- Domain Adaptation and Few-Shot Learning
- Microbial Community Ecology and Physiology
- Oxidative Organic Chemistry Reactions
- Optical Systems and Laser Technology
- Digital Marketing and Social Media
- Human Pose and Action Recognition
- Mass Spectrometry Techniques and Applications
East China Normal University
2024
Hubei University of Technology
2023
ShanghaiTech University
2021-2023
Bridge University
2021
Abstract. Secondary organic aerosols (SOAs) from highly volatile compounds (VOCs) are currently not well represented in numerical models as their heterogeneous formation mechanisms the atmosphere remain unclear. Based on smog chamber experiments, here we investigated yield and pathway of SOA acetone photochemical reactions under low-NOx conditions presence preexisting haze particles ((NH4)2SO4 NH4HSO4) saline mineral (Na2SO4) ammonia-rich conditions. Our results showed that acetone-derived...
In response to challenges in detecting small targets aerial images, such as mutual occlusion, drastic variations target scale, and difficulty localization, this paper presents improvements the YOLOv7 object detection model. Firstly, CBAM fusion attention mechanism is introduced into shallow layers of Backbone, allowing model better focus on regions enhance precision localization. Secondly, ASPP dilated convolutions are integrated SPPCSPC module, increasing receptive field without introducing...
Addressing the challenges of small target feature loss in image detection, susceptibility to background noise interference, and difficulties localization, this paper presents enhancements YOLOv7 detection model. Initially, first CBS layer is pruned, Backbone's SPPCSPC module incorporates self-attention mechanism while shrinking pooling kernel. This modification aims reduce parameter count, redirecting model's focus towards dense small-target regions. Subsequently, integrated into neck...
Scene Graph Generation (SGG) remains a challenging visual understanding task due to its compositional property. Most previous works adopt bottom-up, two-stage or point-based, one-stage approach, which often suffers from high time complexity suboptimal designs. In this paper, we propose novel SGG method address the aforementioned issues, formulating as bipartite graph construction problem. To issues above, create transformer-based end-to-end framework generate entity and entity-aware...
Abstract. Secondary organic aerosols (SOA) from highly volatile compounds (VOCs) are currently not well represented in numerical models as their heterogeneous formation mechanisms the atmosphere remain unclear. Based on smog chamber experiments, here we investigated yield and pathway of SOA acetone photooxidation presence preexisting haze particles ((NH4)2SO4, NH4HSO4) mineral dusts (Na2SO4) under ammonia-rich conditions. Our results showed that acetone-derived can be remarkably enhanced via...
Abstract. Atmospheric isoprene can be oxidizene SOA yield on NOx concentrations was investigated by performing a series of batch chamber experiments; both the gas and aerosol phase chemical species were characterized using High-Resolution Time-of-Flight Chemical Ionization Mass Spectrometer (HR-TOF-CIMS) Aerosol (HR-TOF-AMS), along with an Observation-Based Model (OBM) incorporated Master Mechanism (OBM-MCM model) simulation. We found that could influence formation ultralow volatility...
Abstract Phosphoproteomics integrating data-independent acquisition (DIA) has enabled deep phosphoproteome profiling with improved quantification reproducibility and accuracy compared to data-dependent (DDA)-based phosphoproteomics. DIA data mining heavily relies on a spectral library that in most cases is built DDA analysis of the same sample. Construction this project-specific impairs analytical throughput, limits proteome coverage, increases sample size for Herein we introduce novel...
This work aims to provide a reliable and intelligent prediction model for future trends in sharing economy. Moreover, it presents valuable insights decision-making policy development by relevant governmental bodies. Furthermore, the study introduces predictive system that incorporates an enhanced Harris Hawk Optimization (HHO) algorithm K-Nearest Neighbor (KNN) forecasting framework. The method utilizes improved simulated annealing mechanism Gaussian bare bone structure improve original HHO,...