A5044971247- Remote-Sensing Image Classification
- Remote Sensing and Land Use
- Remote Sensing in Agriculture
- Grouting, Rheology, and Soil Mechanics
- Geotechnical Engineering and Analysis
- Advanced Image and Video Retrieval Techniques
- Metal-Organic Frameworks: Synthesis and Applications
- Infrared Target Detection Methodologies
- Advanced Neural Network Applications
- Advanced Graph Neural Networks
- Metabolomics and Mass Spectrometry Studies
- Digital Media Forensic Detection
- Carbon dioxide utilization in catalysis
- Solidification and crystal growth phenomena
- Horticultural and Viticultural Research
- Virtual Reality Applications and Impacts
- Advanced Steganography and Watermarking Techniques
- Radiation Detection and Scintillator Technologies
- Advanced Semiconductor Detectors and Materials
- Advanced Image Fusion Techniques
- High Temperature Alloys and Creep
- CCD and CMOS Imaging Sensors
- Advanced Materials Characterization Techniques
- CO2 Reduction Techniques and Catalysts
- Ginseng Biological Effects and Applications
Chinese Academy of Sciences
2006-2024
Fujian Institute of Research on the Structure of Matter
2024
University of Chinese Academy of Sciences
2024
Chongqing University of Science and Technology
2024
East China University of Technology
2021-2022
Aerospace Information Research Institute
2021-2022
Ministry of Transport
2022
Ministry of Natural Resources
2022
Universidad Autonoma de Manizales
2021
Wuhan University
2018
Deep learning techniques have boosted the performance of hyperspectral image (HSI) classification. In particular, convolutional neural networks (CNNs) shown superior to that conventional machine algorithms. Recently, a novel type called capsule (CapsNets) was presented improve most advanced CNNs. this paper, we present modified two-layer CapsNet with limited training samples for HSI classification, which is inspired by comparability and simplicity shallower deep models. The trained using two...
Object detection is a focal point in remote sensing applications. Remote images typically contain large number of small objects and wide range orientations across objects. This results great challenges to object approaches based on images. Methods directly employ channel relations with equal weights construct information features leads inadequate feature representation complex image tasks. Multiscale methods improve the speed accuracy detection, while themselves limited information, are...
The nascent graph representation learning has shown superiority for resolving data. Compared to conventional convolutional neural networks, graph-based deep the advantages of illustrating class boundaries and modeling feature relationships. Faced with hyperspectral image (HSI) classification, priority problem might be how convert data into irregular domains from regular grids. In this regard, we present a novel method that performs localized filtering on HSIs based spectral theory. First,...
Abstract A copper porphyrin‐derived metal–organic framework electrocatalyst, FICN‐8, was synthesized and its catalytic activity for CO 2 reduction reaction (CO RR) investigated. FICN‐8 selectively catalyzed electrochemical of to in anhydrous acetonitrile electrolyte. However, formic acid became the dominant RR product with addition a proton source system. Mechanistic studies revealed change major pathway upon addition, while catalyst‐bound hydride (*H) species proposed as key intermediate...
Abstract. Graph-based deep learning has been proved a promising approach that an apparent superiority for graph data and modeling spatial topological relations between features. In particular, attention networks (GATs) are good at efficiently processing the graph-structured hyperspectral by leveraging masked self-attention layers to address known shortcomings of previous frameworks based on convolutions or their approximations. this study, we proposed novel combines localized spectral...
Machine learning-based remote-sensing techniques have been widely used for the production of specific land cover maps at a fine scale. P-learning is collection machine learning training class descriptors on positive samples only. Panax notoginseng rare medicinal plant, which also has highly regarded traditional Chinese medicine resource in China hundreds years. Until now, scarcely observed and monitored from space. Remote sensing natural resources provides us new insights into inventory...
<title>Abstract</title> Using the Chongqing University City double-track tunnel project as a case study, this study comprehensively applied field measurements and numerical simulation methods to analyze impact of over-undercuts during construction on stability using finite element simulation. Intelligent scanning equipment for excavation contours, over-undercut conditions surrounding rock after blasting were measured in detail. The software ANSYS was employed establish stratum structure...
In this short article, we briefly retrospect the recent progress of spectral graph neural networks with manifold-learning-based feature extraction for hyperspectral image classification.
Chinese herbal medicine has played an important role in the treatment of novel coronavirus patients. Machine learning-based remote-sensing techniques play a significant quantitative resource inventory materia medica resources, particularly to explore monitoring abilities for sustainable utilization and biodiversity protection cultivated medicinal plant Panax notoginseng macrocosm. Until now, best knowledge, concrete planting patterns are still poorly known. In this study, two popular...
Hyperspectral imaging is particularly useful for per-pixel thematic classification by unique spectral signatures of landscape materials. Deep learning techniques such as convolutional neural networks have boosted the performance image classification. Recently, several composite learning-based networks, i.e., deep residual (ResNets) and dense (DenseNets), been presented to learn feature representation classification, achieve high accuracies. In this paper, we present a fairly comparable...