A5005905993- Remote-Sensing Image Classification
- Image Retrieval and Classification Techniques
- Advanced Image Fusion Techniques
- Advanced MRI Techniques and Applications
- Advanced Image Processing Techniques
- Rock Mechanics and Modeling
- Medical Imaging Techniques and Applications
- Drilling and Well Engineering
- Geochemistry and Geologic Mapping
- Bee Products Chemical Analysis
- Ginger and Zingiberaceae research
- Bioactive Natural Diterpenoids Research
- Sparse and Compressive Sensing Techniques
- Remote Sensing and Land Use
- Hydraulic Fracturing and Reservoir Analysis
- Face and Expression Recognition
James Cook University
2024-2025
Hainan Provincial Academy of Agricultural Sciences
2025
Animal Science Research Institute
2025
China University of Geosciences
2022
Convolutional Neural Networks (CNN) and Graph (GNN), such as Attention (GAT), are two classic neural network models, which applied to the processing of grid data graph respectively. They have achieved outstanding performance in hyperspectral images (HSIs) classification field, attracted great interest. However, CNN has been facing problem small samples GNN pay a huge computational cost, restrict models. In this paper, we propose Weighted Feature Fusion Network (WFCG) for HSI classification,...
Lithological identification and mapping using remote sensing (RS) imagery are challenging. Traditional lithological relies mainly on multispectral data machine learning methods. However, inadequate spectral information inappropriate classification algorithms major problems for RS geological applications. Moreover, satellite hyperspectral images (HSI) at low spatial resolution convolutional neural network (CNN)-based methods with incomplete feature extraction remain challenging because of the...
Deep learning has achieved impressive results on hyperspectral images (HSIs) classification. Among them, both convolutional neural networks (CNNs) and graph (GNNs) have great potential for image Supervised CNNs can efficiently extract hierarchical spatial-spectral features of images, but these methods face the problem high time complexity as number network layers increases. Semi-supervised GNNs rapidly capture structural information HSIs, while they cannot be well extended to applications...
The genus Mallotus oblongifolius (MO), a member of the Euphorbia family, exhibits predominant distribution in Hainan Island and has been proven to possess diverse medicinal attributes. Research indicates that ultramicro-grinding fully exposes active ingredients oblongifolius, enhancing bioavailability efficacy, compared before. Our study investigates effects ultrafine powder (MOUP) on pigs. A total sixty-four healthy castrated pigs (ternary hybrid pigs, Duroc × Tunchang) with comparable...
Hyperspectral image (HSI) classification techniques have been intensively studied and a variety of models developed. However, these HSI are confined to pocket unrealistic ways datasets partitioning. The former limits the generalization performance model latter is partitioned leads inflated evaluation metrics, which results in plummeting real world. Therefore, we propose universal knowledge embedded contrastive learning framework (KnowCL) for supervised, unsupervised, semisupervised...
Magnetic resonance imaging (MRI) is crucial for enhancing diagnostic accuracy in clinical settings. However, the inherent long scan time of MRI restricts its widespread applicability. Deep learning-based image super-resolution (SR) methods exhibit promise improving resolution without additional cost. Due to lacking aligned high-resolution (HR) and low-resolution (LR) pairs, unsupervised approaches are widely adopted SR reconstruction with unpaired images. these still require a substantial...