A5101666393- Thermochemical Biomass Conversion Processes
- Infrastructure Resilience and Vulnerability Analysis
- Subcritical and Supercritical Water Processes
- Computational Drug Discovery Methods
- Conducting polymers and applications
- Advanced Neural Network Applications
- Perovskite Materials and Applications
- Energy and Environment Impacts
- Underground infrastructure and sustainability
- Integrated Energy Systems Optimization
- Dye analysis and toxicity
- Recycling and Waste Management Techniques
- Advanced MRI Techniques and Applications
- Photoacoustic and Ultrasonic Imaging
- Advanced Image Processing Techniques
- Catalysts for Methane Reforming
- Industrial Vision Systems and Defect Detection
- Lignin and Wood Chemistry
- Coal Combustion and Slurry Processing
- Organic Light-Emitting Diodes Research
- Evacuation and Crowd Dynamics
- Smart Grid Security and Resilience
- Chemical Looping and Thermochemical Processes
- Human Mobility and Location-Based Analysis
- Image and Video Quality Assessment
Chongqing University of Posts and Telecommunications
2023-2024
Everbright International (China)
2022-2023
Shanghai University
2021-2023
Chang'an University
2023
Tongji University
2020
State Key Laboratory of Modern Optical Instruments
2019
Zhejiang University
2019
Wuhan University of Science and Technology
2018
Unmanned aerial vehicle (UAV) technologies combined with digital image processing have been applied to the crack inspection of bridge structures overcome drawbacks manual visual inspection. However, because environmental interference such as uneven natural light, noises produced by UAV hardware, spots on road surface, and jitter, collected images UAVs are usually fuzzy relatively low contrast. In traditional edge detection algorithms Canny algorithm, Prewitt Sobel algorithm accuracy their...
Predicting the formability of perovskite structure for hybrid organic–inorganic perovskites (HOIPs) is a prominent challenge in search required materials from huge space. Here, we propose an interpretable strategy combining machine learning with shapley additive explanations (SHAP) approach to accelerate discovery potential HOIPs. According prediction best classification model, top-198 nontoxic candidates probability (Pf) >0.99 are screened 18560 virtual samples. The SHAP analysis reveals...
In this paper, numerical investigation and optimization is conducted upon an improved updraft gasifier which expected to overcome the weakness of conventional gasifier. The comprehensive Aspen Plus model based on RYield RCSTR reactor. tar prediction constructed, yield determined by volatile biomass gasification temperature. simulation results agree very well with experiment for product yields efficiency, shows accuracy model. content in syngas proved be much lower than that one inflection...
The traditional Sparse Chirp Stepped-frequency ISAR imaging method has the problem of low quality and large computational complexity.In view above problems, this paper studies a sparse based on two-dimensional fast iterative shrinkage threshold algorithm (2D-FISTA).Firstly, echo model signal is constructed.Based model, used to reconstruct result.This can be reconstructed in matrix domain directly, which avoids computation vector processing method, image loss decoupling method.The simulation...
Deep learning has excelled in single-image super-resolution (SISR) applications, yet the lack of interpretability most deep learning-based SR networks hinders their applicability, especially fields like medical imaging that require transparent computation. To address these problems, we present an interpretable frequency division network operates image domain. It comprises a module and step-wise reconstruction method, which divides into different frequencies performs accordingly. We develop...
In recent years, instruction-based image editing methods have garnered significant attention in editing. However, despite encompassing a wide range of priors, these are helpless when handling tasks that challenging to accurately describe through language. We propose InstructBrush, an inversion method for bridge this gap. It extracts effects from exemplar pairs as instructions, which further applied Two key techniques introduced into Attention-based Instruction Optimization and...
A demura method using high resolution monochrome camera is proposed for OLED panels, enabling single data capturing white images under different graylevels (GLs). Our minimizes TACT low GLs due to its efficiency of light utilization, compared with color Bayer pattern.
This paper proposes a method implementing manifold learning techniques to accomplish simultaneous reconstruction and segmentation of brain MRI. Simultaneous by can not only overcome the conflict efficiency precision, but also concentrate on region interest. In our study, we present unified framework for image based data-driven supervised task that allows mapping between sensor domain. The module was consisted reconstructing phase segmenting which pre-trained ImageNet, MGH-USC HCP Neonatal...
Two-dimensional (2D) organic-inorganic hybrid perovskites (OIHPs) have drawn increased attention due to rich physical properties such as ferroelectricity and photovoltaic properties. Nevertheless, it is challenging discover novel 2D OIHPs based on that favors good photoelectric conversion efficiency. Herein, a diagrammatic machine learning model was employed improve this issue. We collected 179 with variety of organic cations screened out 6 features from 10622 descriptors. Subsequently,...