A5061973097- Advanced Image Processing Techniques
- Advanced Vision and Imaging
- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Grouting, Rheology, and Soil Mechanics
- Geotechnical Engineering and Analysis
- Advanced Decision-Making Techniques
- Image Enhancement Techniques
- Mobile Agent-Based Network Management
- Cancer-related molecular mechanisms research
- Geoscience and Mining Technology
- Image Processing and 3D Reconstruction
- Sleep and Wakefulness Research
- Image and Signal Denoising Methods
- Mobile Ad Hoc Networks
- Remote Sensing and Land Use
- Fire Detection and Safety Systems
- Engineering and Test Systems
- Experimental Learning in Engineering
- Video Surveillance and Tracking Methods
- Tunneling and Rock Mechanics
- Retinal Diseases and Treatments
- Information and Cyber Security
- Geotechnical Engineering and Underground Structures
- Retinal Imaging and Analysis
Eye & ENT Hospital of Fudan University
2024
State Key Laboratory of Medical Neurobiology
2024
Zhejiang Provincial People's Hospital
2024
PLA Army Engineering University
2012-2023
Lanzhou University
2021
Huawei Technologies (Sweden)
2021
Tsinghua University
2021
China Meteorological Administration
2020
Beijing Shijingshan Hospital
2009-2010
This paper presents a new meteorological photo classification system based on the Multi-channel Convolutional Neural Network (CNN) and improved Frame Difference Method (FDM). can work in an embedded with limited computational resources categorize cloud observation photos captured by ground cameras. We propose FDM extractor to detect extract cloud-like objects from large into small images. Then, these images are sent CNN image classifier. construct classifier train it photo-set that we...
Deep learning based methods, especially convolutional neural networks (CNNs) have been successfully applied in the field of single image super-resolution (SISR). To obtain better fidelity and visual quality, most existing are heavy design with massive computation. However, computation resources modern mobile devices limited, which cannot easily support expensive cost. this end, paper explores a novel frequency-aware dynamic network for dividing input into multiple parts according to its...
Command protection engineering is the important component of national system. To raise level its construction, a deformation prediction model given based on Genetic Algorithm (GA), Least Square Support Vector Machines (LSSVM) and markov theory. algorithm used to improve parameter LSSVM. Markov predict method precision model. Finally, be certain command engineering, accuracy improved obviously. The proved credible precise.
Over the past decades, feature-based statistical machine learning and deep neural networks have been extensively utilized for automatic sleep stage classification (ASSC). Feature-based approaches offer clear insights into characteristics require low computational power but often fail to capture spatial–temporal context of data. In contrast, can process raw signals directly deliver superior performance. However, their overfitting, inconsistent accuracy, cost were primary drawbacks that...
Ultrawidefield fundus (UWF) images have a wide imaging range (200° of the retinal region), which offers opportunity to show more information for ophthalmic diseases. Image quality assessment (IQA) is prerequisite applying UWF and crucial developing artificial intelligence-driven diagnosis screening systems. Most image systems been applied assessments natural images, but whether these are suitable evaluating remains debatable. Additionally, existing IQA datasets only provide photographs...
This paper illustrated the characteristics of UTP-5 and Hub-spoken topology, presented concept margin motion coefficient reliability based on graph theory relation matrix made thorough analysis comparison. Suggestions are resolution topology problems cabling infrastructure intelligent buildings.
Abstract The geological conditions are complex in the Longnan mountain area, and landslide disasters widely distributed numerous, among which large accumulation layer is especially developed. This kind of mostly along river basin has characteristics multi-phase multistage. slide body state creep plastic flow, it easy to revive. In this paper, evolution model secondary were studied based on field survey numerical calculation method, order provide reference for monitoring treatment area....
When the tunnel is excavated through water-rich fault, it easy to cause problems of water surge and instability face tunnel. The advanced curtain grouting can effectively improve rock mass quality permeability fault broken zone, ensure safety construction. In view this, MADIS-GTS finite element software used establish dynamic excavation models under conditions no grouting. deformation seepage characteristics palm surface were compared analysed determine fortification length, optimal...
This paper describes a idea of the new E-Commerce network framework and its implementation in virtue platform composed Hot-Standby Two-Node-Cluster, Virtual Server, related software. research focuses on how to enhance security data storage, processing, transmission exchange.
This article has been retracted by the publisher.
Deep learning based methods, especially convolutional neural networks (CNNs) have been successfully applied in the field of single image super-resolution (SISR). To obtain better fidelity and visual quality, most existing are heavy design with massive computation. However, computation resources modern mobile devices limited, which cannot easily support expensive cost. this end, paper explores a novel frequency-aware dynamic network for dividing input into multiple parts according to its...
Abstract In the engineering fact that excavation will lead to deterioration of rock parameters in damage zone tunnel, based on a diversion tunnel by simulating continuous dynamic adjustment process surrounding parameters, law deep buried soft is studied. It considered damaged area constructed TBM process. After excavation, angle friction φ increases, cohesive strength C decreases, and supporting structure have limiting effect parameters.