A5065727796- Remote-Sensing Image Classification
- Economic theories and models
- Monetary Policy and Economic Impact
- Advanced Image Fusion Techniques
- Economic Theory and Policy
- Advanced Algorithms and Applications
- Advanced Image and Video Retrieval Techniques
- Remote Sensing and Land Use
- Non-Destructive Testing Techniques
- Fiscal Policy and Economic Growth
- Spectroscopy and Chemometric Analyses
- Advanced Image Processing Techniques
- Image Processing Techniques and Applications
- Automated Road and Building Extraction
- Iterative Learning Control Systems
- Leaf Properties and Growth Measurement
- Infrastructure Maintenance and Monitoring
- Hydraulic and Pneumatic Systems
- Banking stability, regulation, efficiency
- Smart Agriculture and AI
- Image Retrieval and Classification Techniques
- Cavitation Phenomena in Pumps
- Cell Image Analysis Techniques
- Modeling, Simulation, and Optimization
- Industrial Vision Systems and Defect Detection
China Southern Power Grid (China)
2024
China Agricultural University
2024
Henan Normal University
2024
Zhengzhou University of Light Industry
2023
Henan University
2021
Chongqing University
2021
Fudan University
2018
East Carolina University
2011-2013
Deep convolutional networks perform well in remote sensing (RS) image classification. Usually, it is difficult to obtain a large number of labeled samples classification tasks. Traditionally, the acquisition images quite different from photos provided by digital cameras. However, imaging system for high resolution (HR) RS (often with RGB 3 channels) similar those In paper, transfer learning algorithm based on deep neural proposed attack problem lacking samples, particular context pre-trained...
Convolutional neural networks (CNNs) and graph convolutional (GCNs) have led to promising advancements in hyperspectral image (HSI) classification; however, traditional CNNs with fixed square convolution kernels are insufficiently flexible handle irregular structures. Similarly, GCNs that employ superpixel nodes instead of pixel may overlook pixel-level features; both tend extract features locally cause loss multilayer contextual semantic information during feature extraction due the kernel....
A hyperspectral image (HSI) classification algorithm that combines graph convolutional networks (GCNs) and neural (CNNs) aims to generate complementary spatial-spectral joint information at the superpixel pixel levels. However, CNN part is typically a single 2D or 3D network cannot fully capture middle long-range spatial relationships between pixels. Additionally, GCNs commonly under-segmented in segmentation process does not consider weight neighboring superpixels when calculating adjacency...
An innovative framework for peach tree disease recognition and segmentation is proposed in this paper, with the aim of significantly enhancing model performance complex agricultural settings through deep learning techniques data fusion strategies. The core innovations include a tiny feature attention mechanism backbone network, an aligned-head module, Transformer-based semantic specially designed alignment loss function. integration these technologies not only optimizes model’s ability to...
In recent decades, it is easy to obtain remote sensing images which have been successfully applied various applications, such as urban planning, hazard monitoring, etc. particular, high resolution (HR) (RS) can better monitor our living environment from a broader spatial perspective. However, raw provide no labeling information train classifier, usually exploited generate maps. Based on previous work, in the paper, an automatic classification system proposed classify RS using deep neural...
The original Hyperspectral image (HSI) has different degrees of Hughes phenomenon and mixed noise, leading to the decline classification accuracy. To make full use spatial-spectral joint information HSI improve accuracy, a novel dual feature extraction framework transform domain-spatial domain filtering based on multi-scale-superpixel-dimensionality reduction (LRS-HRFMSuperPCA) is proposed. Our uses low-rank structure sparse representation repair unobserved part caused by noise then denoises...
Both the maturity structure and liquidity constraints are important to have time-consistent optimal fiscal policy. The of initial bond holding position introduces asymmetry/symmetry in treating consumption goods across time, leading time-inconsistent/consistent In a closed economy, Ramsey governments can treat all symmetrically satisfied. Therefore, policy is time-consistent. small open result an endogenous determinacy restriction from labour market. This on Lagrange multipliers general...
research-article Advanced-ExtremeNet: Combined with Depthwise Separable Convolution for the Detection of Steel Bars Share on Authors: Shuyang Pang Dept. Big Data Business, CISDI Information Technology Co., Ltd, China ChinaView Profile , Xuan Liu Chongqing University, Shangwei Mao Hongsheng Jia Bin Authors Info & Claims ICAIIS 2021: 2021 2nd International Conference Artificial Intelligence and SystemsMay Article No.: 156Pages 1–6https://doi.org/10.1145/3469213.3470359Online:18 August...
This paper quantitatively analyzes the impact of money stock on optimal monetary and fiscal policy in a stochastic production economy with sticky prices. The numerical results indicate that sufficient large quantity makes noticeable difference many aspects policy. They suggest volatile inflation China may not be as bad existing theory would have implied if its amount is taken into consideration.