A5102889415- Image and Signal Denoising Methods
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Caching and Content Delivery
- Advanced Computational Techniques and Applications
- Advanced Algorithms and Applications
- Network Security and Intrusion Detection
- Smart Grid and Power Systems
- Power Systems and Technologies
- Advanced Image Fusion Techniques
- Mathematical Analysis and Transform Methods
- Peer-to-Peer Network Technologies
- Mobile Ad Hoc Networks
- Intracerebral and Subarachnoid Hemorrhage Research
- Chinese history and philosophy
- Advanced Data Compression Techniques
- High-Voltage Power Transmission Systems
- Advanced Sensor and Control Systems
- Cloud Computing and Resource Management
- Landslides and related hazards
- Laser Design and Applications
- Spectroscopy and Laser Applications
- Software-Defined Networks and 5G
- Distributed and Parallel Computing Systems
- Nuclear reactor physics and engineering
- Cooperative Communication and Network Coding
Army Medical University
2021-2024
Southwest Hospital
2021-2024
Beijing University of Posts and Telecommunications
2011-2024
Barro Colorado Island
2024
Nanjing University
2024
Lanzhou University
2010-2022
Jiangsu Institute of Parasitic Diseases
2021
Evangelisches Klinikum Niederrhei
2020
Central South University
2020
Third Xiangya Hospital
2020
Decompressive craniectomy (DC), a surgery to remove part of the skull and open dura mater, maybe an effective treatment for controlling intracranial hypertension. It remains great interest elucidate whether DC is beneficial intracerebral hemorrhage (ICH) patients who warrant clot removal (CR) prevent
The artificial intelligence calculation method can effectively solve various nonlinear mapping relationships. strength of these solvers is exploited for the evaluation power grid investment risk using back propagation (BP) neural network optimized by genetic algorithm. mathematical model problem constructed selecting transfer function and defining fitness BP has good ability self-learning, self-adaptation generalization, which overcome drawbacks traditional methods relying on experts'...
Purpose The study aims to build a high-precision longitudinal dynamics model for heavy-haul trains and validate it with line test data, present an optimization method multi-stage cyclic brakes based on the conduct multi-objective detailed evaluation of driver’s manipulation during braking. Design/methodology/approach train was established verified by braking data 20,000 t combination long steep downgrade. Then genetic algorithm is employed subsequent decoupling multiple procedures, due...
Previous studies have demonstrated abnormal local spontaneous brain activity in the conventional frequency bands (0.01-0.08 Hz) obstructive sleep apnea (OSA). However, it is not clear whether these abnormalities are associated with specific band of low-frequency oscillations or can be improved a continuous positive airway pressure (CPAP) treatment. This study aimed to investigate regional homogeneity (ReHo) at baseline (pre-CPAP) and after one month CPAP adherence treatment (post-CPAP) OSA...
Microgrid, as a distributed power technology, has deep potential at present. This study deeply researches microgrid and electric vehicles. A grid-connectedmicrogrid optimization management model is established, an adaptive crossover multi particle swarm (ACM-PSO) proposed, which can adjust operate individually. The proposed algorithm introduces individual adjustment operation considers the start stop state of micro supply. Besides, it does not need to decompose unit commitment problem into...
In this paper, we propose an improved rotation invariant uniform local binary pattern (RIU-LBP) operator for segmenting high-resolution sensing image which can effectively describe the texture features of a remote image. The RIU-LBP is based on RIU-LBP. It introduces threshold in binarization region pixels. new LBP better tolerate small variation and distinguish plain rough than original does. Then, merging criterion regions proposed, regional value distribution Bhattacharyya distance....
Differing from traditional semi-supervised learning, class-imbalanced learning presents two distinct challenges: (1) The imbalanced distribution of training samples leads to model bias towards certain classes, and (2) the unlabeled is unknown potentially that labeled samples, which further contributes class in pseudo-labels during training. To address these dual challenges, we introduce a novel approach called Twice Class Bias Correction (TCBC). We begin by utilizing an estimate...
Currently, the Internet has become an significant part of global communication infrastructure. With more and new services emerging, current IP network may not satisfy user demands in future. At present, most routing scheme typically routes all traffic over one discovered single path, which are inherently slow responding to congestion burst. Furthermore, use only path often causes waste resources. As a single-path protocol, Open Shortest Path First (OSPF) suffers same problems. An alternative...
In this paper, a SAR target recognition method is proposed based on the improved joint sparse representation (IJSR) model. The IJSR model can effectively combine multiple-view images from same physical to improve performance. classification process contains two stages. Convex relaxation used obtain support sample candidates with ℓ1-norm minimization in first stage. low-rank matrix recovery strategy introduced explore final samples and its corresponding coefficient second Finally, minimal...
This paper presents a hybrid clustering algorithm based on density and ant colony algorithm, that to determine the initial cluster centers according objects distribution method, then use swarm intelligence randomness of find arbitrary shape clusters, avoid falling into local convergence, get relatively stable global optimal solution. Theoretical analysis experimental results show improved can achieve better results.
The radiation shadow shield of the space nuclear power is an important component to protect electronic device at dose plane. In this paper, design first modeled as a multi-objective optimization problem, and then converted into single objective problem by setting other sub-objectives constraints. An method combining genetic algorithm with Monte Carlo simulation developed solve design. Through optimization, weight can be greatly reduced (14.7% initial 16.4% best individual generation), while...
Effective noise removal has become a hot topic in image denoising research while preserving important details of an image. An adaptive threshold algorithm based on fitting diffusion is proposed. Firstly, the coefficient equation improved, and established to overcome defects texture detail loss edge degradation caused by excessive intensity. Then, function adaptively designed improved so that it can automatically control according maximum gray value number iterations, as further preserve such...
To solve the existing problems of traditional bridge vibration (BV) data compression methods, on basis deep analysis for wavelet transform (WT), a method combined with LZW coding is proposed in this paper. In accordance characteristics and properties transform-based multi-resolution analysis, Donoho threshold algorithm adopted to quantify high-frequency decomposed components. Then are compressed improved coding. justify method, several groups actual sampling simulated. The results indicate...
In this paper, a clustering algorithm based on the immune mechanism of capture antigen by antibody has been presented. The datum to be clustered are viewed as antigens,and cluster centers antibodies in system. is effectively process which system constantly generates for recognition antigens and finally optimal antigens. experimental results show that can successfully applied data clustering.
Two extraction models applied in estimating forest height and above-ground biomass (AGB) were developed using the X-band Interferometric Synthetic Aperture Radar (InSAR) data, which was acquired from China airborne SAR system 2013 covering part of forested area northeast China. The multi-passes InSAR data for estimations AGB introduced respectively with detailed ground true derived both plot Light Detection And Ranging (LiDAR) data. estimation results indicated that it is feasible to...
The radiometric quality of polarimetric SAR (PolSAR) image is affected by terrain undulations due to 1) the variation effective scattering area, 2) mechanisms, and 3) polarization states. This paper proposed a three-stage correction method PolSAR data for impact above three aspects. PALSAR-2 LiDAR forest AGB are used illustrate method. results showed that effects can been effectively removed correlation between biomass backscatter coefficient improved after correction. HV has best with (R =...