A5100644863- Perovskite Materials and Applications
- Advanced Photocatalysis Techniques
- Advanced Image Processing Techniques
- Polyoxometalates: Synthesis and Applications
- Advanced Decision-Making Techniques
- Visual perception and processing mechanisms
- Robotic Locomotion and Control
- Luminescence and Fluorescent Materials
- Topic Modeling
- Cancer-related molecular mechanisms research
- Video Coding and Compression Technologies
- Software Reliability and Analysis Research
- Advanced Vision and Imaging
- Acne and Rosacea Treatments and Effects
- Image and Object Detection Techniques
- Higher Education and Teaching Methods
- Reinforcement Learning in Robotics
- Image Enhancement Techniques
- Non-Invasive Vital Sign Monitoring
- Geochemistry and Geologic Mapping
- Software Engineering Research
- Cardiac Health and Mental Health
- Image and Signal Denoising Methods
- Heart Failure Treatment and Management
- Expert finding and Q&A systems
Soochow University
2025
ShanghaiTech University
2024
Soochow University
2024
Chengdu Technological University
2023
Qingdao University of Technology
2023
Jiangsu University of Technology
2015-2022
Chengdu Second People's Hospital
2022
Chongqing Science and Technology Commission
2022
Chongqing Medical University
2022
Army Medical University
2022
A recent study namely software defect prediction model based on Local Linear Embedding and Support Vector Machines (LLE-SVM) has indicated that Regression (SVR) an interesting potential in the field of prediction. However, parameters optimization LLE-SVM is computationally expensive by using grid search algorithm, resulting a lower efficiency model; it ignores imbalance data sets when SVM classier to differentiate defective class non-defective class. Thus accuracy. To solve these problems...
Current mainstream super-resolution algorithms based on deep learning use a convolution neural network (CNN) framework to realize end-to-end from low-resolution (LR) image high-resolution (HR) images, and have achieved good restoration effects. However, as the number of layers in is increased, better results are not necessarily obtained, there will be problems such slow training convergence, mismatched sample blocks, unstable results. We propose preclassified deep-learning algorithm...
Images taken in extremely low light suffer from various problems such as heavy noise, blur, and color distortion. Assuming the low-light images contain a good representation of scene content, current enhancement methods focus on finding suitable illumination adjustment but often fail to deal with noise Recently, some works try suppress reconstruct raw data. But these apply network instead an image signal processing pipeline (ISP) map data enhanced results which leads learning burden for get...
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
Patients with heart failure preserved ejection fraction (HFpEF) are more often female, but gender differences in psychological distress patients HFpEF have not been determined.We aimed to compare anxiety, depression, insomnia, and quality of life (QoL) between women men HFpEF.A total 263 consecutive hospitalized were enrolled a multicenter study. Demographic clinical characteristics recorded. Anxiety depression assessed using the Hospital Depression Scale (HADS), insomnia was by Insomnia...
Abstract Squatting is a basic movement of bipedal robots, which essential in robotic actions like jumping or picking up objects. Due to the intrinsic complex dynamics perfect squatting motion requires high‐performance planning and control algorithms. The standard academic solution combines model predictive (MPC) with whole‐body (WBC), usually computationally expensive difficult implement on practical robots limited computing resources. real‐time kinematic prediction (RKP) method proposed,...
Digital image processing technology is widely used in the further application of computer graphics.Based on analysis characteristics digital course, instructional objectives, project based learning theory and method was applied to reform practice experimental processing.Based necessity teaching reform, a practical engineering application, which including four parts such as designing components, organizing continents, practicing after class evaluating students proposed.On basis target set...
For issue that coal mining data tends to be incomplete, noisy and inconsistent, some popular classifiers are applied predict class label of dataset.Noise bad points rejected from which will exchanged input format suitable for mining.Then different used classify after extracting features.In the end, classification results analyzed knowledge assimilation is done.Experiment show decision tree model gives 88% accuracy correctly whereas neural network predicts 85% correct label.This research...
Rapid Propagation of malicious program has caused great harm to the security user information, traditional way killing methods, which is lagging behind and nonintelligent, been unable meet demand current detection.Studying new detection method on Windows Platform, screening out intelligent rules model feature executable extracting methods based data mining.Introducing sample processing selection process, analyzing simulating classification method, result shows that can effectively improve...
Studies show that encoding technologies in H.264/AVC, including prediction and conversion, are essential technologies. However, these more complicated than the MPEG-4, which is a standard method widely adopted worldwide. Therefore, amount of calculation H.264/AVC significantly up-regulated compared to MPEG-4. In present study, it intended simplify computational expenses international compression coding system for moving images. Inter refers most feasible technology, taking up 60% entire...
Abstract In order to improve the detection accuracy for small-scale targets in complex scenes, an improved target algorithm based on EfficientNet is proposed. Firstly, network used optimize DarkNet53 feature extraction network. Compressing standard convolution with depthwise separable convolution, and increasing depth of neural residual that can effectively achieve extraction, reduce number parameters speed. Secondly, a pyramid design four scales features multi-scale which improves small...
Geochemical anomalies are the basis of mineral deposit prediction. Through study characteristics geochemical anomalies, we found that their distribution was consistent with a generalized Pareto (GPD). In present study, designed model for anomaly extraction via GPD. GPD model, used kurtosis method to estimate threshold value Furthermore, principal component analysis (PCA) extract comprehensive information different elements in which minerals enriched. On this basis, new algorithm named GPDA...
The use of noise has been a powerful tool in learning about nonlinear stages visual processing. We have extended the approach Petrov, Dosher & Lu (2005, 2006) to enable us gather more accurate data and extend their analysis manner that enables isolate Petrov measured Gabor orientation discrimination periphery with added oriented noise. Observers reported target (left/right) for several contrasts presence background was congruent or incongruent orientation. In our experiments following...
With the development of technology, technology and equipment for collecting physical signs are gradually maturing, data analysis ability medical treatment also developing rapidly. However, few models can analyze in real-time conclude, then transmit them to emergency centers hospitals. Therefore, a warning automatic call help model based on vital were constructed. Based finite automata, took as condition state transition. To ensure accuracy data, 448,972 triage table MIMIC-IV database used...