A5080647183- Remote-Sensing Image Classification
- Advanced Image and Video Retrieval Techniques
- Advanced Neural Network Applications
- Statistical Methods and Inference
- Bayesian Methods and Mixture Models
- Power System Reliability and Maintenance
- Electrical Fault Detection and Protection
- Automated Road and Building Extraction
- Medical Image Segmentation Techniques
- Advanced Computational Techniques and Applications
- Infrastructure Maintenance and Monitoring
- Advanced Battery Materials and Technologies
- Advanced Battery Technologies Research
- Perovskite Materials and Applications
- Anomaly Detection Techniques and Applications
- Advanced Image Fusion Techniques
- Network Security and Intrusion Detection
- Land Use and Ecosystem Services
- Markov Chains and Monte Carlo Methods
- Asphalt Pavement Performance Evaluation
- Advanced Decision-Making Techniques
- Advanced Vision and Imaging
- Gaussian Processes and Bayesian Inference
- Stochastic Gradient Optimization Techniques
- Concrete Corrosion and Durability
Lanzhou Jiaotong University
2023-2025
North China Electric Power University
2023
Henan Normal University
2020
Dalian Jiaotong University
2017
Dalian University
2017
Peking University
2017
Zhejiang University of Science and Technology
2012
Xi'an University of Science and Technology
2009
In recent years, with the gradual increase of neural network Params (the aggregate trainable elements in a model, including weights, biases, and other adjustable elements) calculation volume, model compression within an acceptable range accuracy variations has emerged as prominent research focus field deep learning. Model pruning knowledge distillation have been widely used for reducing complexity storage cost networks. This study designs Remote Sensing Object Detection Network (RSODNet),...
In this paper we propose a novel framework for the construction of sparsity-inducing priors. particular, define such priors as mixture exponential power distributions with generalized inverse Gaussian density (EP-GIG). EP-GIG is variant hyperbolic distributions, and special cases include scale mixtures Laplace mixtures. Furthermore, can subserve Bayesian sparse learning nonconvex penalization. The densities be explicitly expressed. Moreover, corresponding posterior distribution also follows...
ABSTRACTLand Use/Land Cover (LULC) classification has become increasingly important in various fields, including ecological and environmental protection, urban planning, geological disaster monitoring. With the development of high-resolution remote sensing satellite technology, there is a growing focus on achieving precise LULC classification. However, accuracy fine-grained challenged by high intra-class diversity low inter-class separability inherent images. To address this challenge, paper...
We demonstrate an innovative strategy for quantum dot-induced improved performance of cadmium telluride solar cells without a Cu buffer layer.
The pavement is vulnerable to damage from natural disasters, accidents and other human factors, resulting in the formation of cracks. Periodic monitoring can facilitate prompt detection repair diseases, thereby minimizing casualties property losses. Due presence numerous interferences, recognizing highway cracks complex environments poses a significant challenge. Nevertheless, several computer vision approaches have demonstrated notable success tackling this issue. We employed novel approach...
Remote sensing images are characterized by complex feature backgrounds and large target scale differences, so object detection for remote is a challenging problem. This work proposes one-stage structure image model called GODANet. First, the GODANet incorporates Global Context Network (GCNet) in extraction structure. The GCNet focuses on region of interest from global perspective. Second, output layer utilizes an omni-dimensional dynamic convolution technique, allowing more flexible...
Three-dimensional geological modeling (3DGM) is a complicated project in the research of three-dimensional geographic information system, and there have many data model for by now. The thesis introduces kind object-oriented model: Component-based Topological Data Model (CTDM). Based on principles analysis, author discusses gives out steps, especially spatial section location, coordinates conversion, Component definition volume representation from this model. In end paper, an example was...
Li6PS5Cl possesses high ionic conductivity and excellent interfacial stability to electrodes is known as a promising solid-state electrolyte material for all-solid-state batteries (ASSBs). However, the optimal annealing process of has not been studied systematically. Here, Box–Behnken design used investigate interactions heating rate, temperature, duration optimize conductivity. The response surface methodology with regression analysis employed simulating data obtained, optimized parameters...
In this paper we propose a novel framework for the construction of sparsity-inducing priors. particular, define such priors as mixture exponential power distributions with generalized inverse Gaussian density (EP-GIG). EP-GIG is variant hyperbolic distributions, and special cases include scale mixtures Laplace mixtures. Furthermore, can subserve Bayesian sparse learning nonconvex penalization. The densities be explicitly expressed. Moreover, corresponding posterior distribution also follows...
The abnormal detection of the workload sequence is designed to achieve intelligent operation and management cloud platform improve operational efficiency. Due diversity variation patterns in large-scale cloud, it difficult for traditional methods extract features effectively, which leads detecting anomalies inaccurately. In this paper, a deep unsupervised anomaly model with fusion spatial temporal (TS-DeepSVDD) proposed. To sequence, introduces convolutional recurrent neural network (CRNN)...
This paper introduces a new method to calculate the weight of influence factors on cable failure in 10kV network. A large number statistical data recent years are investigated analyze main causes and detail. Different from traditional methods calculating factors, methodology introduced this eliminates subjective judgement by using data. Based these data, information theory is employed as objectively factors. In theory, entropy (IE) reflects degree uncertainty. Therefore, above, IE status can...