A5078845147- Advanced Neural Network Applications
- Human Mobility and Location-Based Analysis
- Industrial Vision Systems and Defect Detection
- Land Use and Ecosystem Services
- Advanced Image and Video Retrieval Techniques
- Geological Modeling and Analysis
- Advanced Computational Techniques and Applications
- Advanced Measurement and Detection Methods
- Color perception and design
- 3D Surveying and Cultural Heritage
- Geographic Information Systems Studies
- Reservoir Engineering and Simulation Methods
- Digital Media and Visual Art
- Rice Cultivation and Yield Improvement
- Remote Sensing and Land Use
- Remote Sensing and LiDAR Applications
- Culinary Culture and Tourism
- Image Enhancement Techniques
- Advanced Technologies in Various Fields
- Diverse Aspects of Tourism Research
- Photosynthetic Processes and Mechanisms
- Image and Object Detection Techniques
- Color Science and Applications
- Sustainable Urban and Rural Development
- Biological and pharmacological studies of plants
Huzhou Central Hospital
2025
Sichuan University
2024
Jiangxi Normal University
2020-2024
Anhui University of Science and Technology
2013-2024
University of Jinan
2021-2023
Xinjiang University
2023
Xi'an Jiaotong University
2022
Herzen University
2019
Zaozhuang Municipal Hospital
2019
Anhui Science and Technology University
2013
Multi-object pedestrian tracking plays a crucial role in autonomous driving systems, enabling accurate perception of the surrounding environment. In this paper, we propose comprehensive approach for tracking, combining improved YOLOv8 object detection algorithm with OC-SORT algorithm. First, train model on Crowdhuman dataset detection. The integration advanced techniques such as softNMS, GhostConv, and C3Ghost Modules results remarkable precision increase 3.38% an mAP@0.5:0.95 3.07%....
“A picture is worth a thousand words”. Analysis of the visual content tourist photos an effective way to explore image destinations. With development computer deep learning and big data mining technology, identifying massive numbers by convolutional neural network (CNN) approaches breaks through limitations manual photos’ information, e.g., small sample size, complex identification process, results deviation. In this study, 531,629 travel Jiangxi were identified as 365 scenes technology....
Due to the complexity of wetland ecosystems, wetlands have a wide area alternating land and water zones complex vegetation composition, making it challenging achieve dynamic displays virtual scenes using three-dimensional modeling. This study proposes workflow game engine-based scene construction for rapid modeling scenes. The work utilized Poyang Lake as primary research area. It integrated unmanned aerial vehicle data collection technology geographic information with 3D (three-dimensional)...
Abstract The aim of this study was to examine the prevalence multidrug-resistant tuberculosis (MDR-TB) within (TB) population in Huzhou City, identify associated risk factors, furnish reference values for clinical practice, and establish standardized anti-TB treatment regimens. Retrospectively analyzing data from TB patients City encompassed 2,261 cases with sputum smear positive 1,754 negative January 2016 December 2022. There were 3.66%(147/4,015) identified as MDR-TB. Multivariate...
Access to real-time environmental sensing data is key representing changes in the digital twin of lakes. The visualization an important part establishing a In past, display methods were either charts or two-dimensional map-based methods. Breaking through traditional and realizing multi-dimensional multi-view lakes something that this particular paper tries resolve. This study proposes framework integrate, manage, analyze, visualize addition, also seeks realize coupling expression geospatial...
Bletilla striata (Thunb.) Reichb.f., is a traditional Chinese medicine, and the polysaccharide (BSP) one of principal components extracted from with various biological activities. Previous studies have shown that many natural polysaccharides significant immunomodulatory However, as plant polysaccharide, research BSP on activities limited. In this study, we aim to investigate effect in vivo further explore its underlying mechanism vitro. vivo, cyclophosphamide (CTX)-induced immunosuppression...
In this study, changes in the level of thiol expression on cell surface were utilized both healthy and diseased states. Oral nano-drug delivery system mediated by lipoic acid can be used to treat diabetes inflammatory diseases.
An urban green space network provides safe and exercise routes for residents. This study selected Nanchang as the area. Using fitness application data, we explored mobility of people exercising in network, i.e., flow exercise. Spatial analysis based on social networks, GIS, gravity model was used to analyze nodes, characteristics, mechanism The results show that there were differences hierarchy attractiveness nodes. Distance had an important influence It found walkers moved between adjacent...
Understanding the evolutionary characteristics of innovation network structure can improve urban and regional construction. Urban innovative development is affected by various factors, which be analyzed via models networks. We establish a multi-criteria evaluation system capability use an improved gravity model to construct for 2015–2018, employing social methods analyze structural spatial patterns. Results show that: (1) The cities in agglomeration middle reaches Yangtze River has gradually...
Rich observation data generated by ubiquitous sensors are vital for wetland monitoring, spanning from the prediction of natural disasters to emergency response. Such use different acquisition and description methods and, if combined, could provide a comprehensive wetland. Unfortunately, these remain hidden in isolated silos, their variety makes integration interoperability significant challenge. In this work, we develop semantic model monitoring using an agile modular approach, namely,...
The ceramic industry has been vital to the city’s development and prosperity in Jingdezhen, but of ceramics Jingdezhen unclear since China’s economic reforms, which will become a bottleneck limiting sustainable city. This study explored spatial agglomeration spatiotemporal evolution from 1980 2020 using enterprise directory data. opted for microscopic perspective employed kernel density estimation exploratory data analysis obtain necessary results. It also analyzed influencing factors...
Multiscale virtual environments (MSVEs) allow the integration of elements and at different scale levels into a unified space, which facilitates researchers’ perception, understanding, experimental research complex geospatial spaces. Although there have been several methods for achieving multiscale effects in (VEs), they cannot assist users constructing more complete spatial cognitive maps presenting information efficiently. This study proposes hierarchical-structure-based MSVE construction...
Existing image-text modality alignment in Vision Language Models (VLMs) treats each text token equally an autoregressive manner. Despite being simple and effective, this method results sub-optimal cross-modal by over-emphasizing the tokens that are less correlated with or even contradictory input images. In paper, we advocate for assigning distinct contributions based on its visual correlation. Specifically, present contrasting image inputs, difference prediction logits provides strong...
The newly discovered Liuzhangshan (LZS) Rb deposit is a large granite-type in the West Qinling Orogenic Belt (WQOB), spatially associated with biotite monzogranite. Previous research has focused on geological characteristics and metallogenic conditions of LZS deposit; however, its geochronology petrogenesis remain unclear. In this study, monazite U-Pb dating monzogranite yielded age 200.8 ± 0.9 Ma via laser-ablation inductively coupled plasma mass spectrometry (LA-ICP-MS). Combined magmatic...
The existing object detection algorithms have the problems of poor performance on piping and instrumentation diagram(P&ID) symbol dataset. Therefore, this paper proposes a algorithm in P &ID images using improved network YOLOv5. Firstly, due to limited number samples dataset class imbalance problem, Random Splicing suitable for images. It can selectively augment one or more classes symbols P&ID Then, address problem subtle differences difficulty dis-tinguishing between symbols, introduces...
This paper expects to get the real-time changes of traffic flow by researching relationships among three basic parameters flow, speed and occupancy. we firstly carries out statistical correlation analysis grey relational study connection between speed, occupancy a conclusion that there doesn’t exist significant linear relationship neither comparisons, influence on is little bigger than Then try establish binary regression model without intercept GM (0,3) do data fitting, final simulation...