A5088673166- Image and Signal Denoising Methods
- Advanced Image Fusion Techniques
- Functional Brain Connectivity Studies
- Remote Sensing and Land Use
- Sparse and Compressive Sensing Techniques
- Remote-Sensing Image Classification
- Remote Sensing in Agriculture
- EEG and Brain-Computer Interfaces
- Medical Image Segmentation Techniques
- Traditional Chinese Medicine Studies
- Land Use and Ecosystem Services
- Rangeland Management and Livestock Ecology
- Advanced MRI Techniques and Applications
- Cancer-related molecular mechanisms research
- Tensor decomposition and applications
- Advanced Neuroimaging Techniques and Applications
- Ombudsman and Human Rights
- Advanced Image Processing Techniques
- Algal biology and biofuel production
- Ginseng Biological Effects and Applications
- Face and Expression Recognition
- Impact of AI and Big Data on Business and Society
- Image Enhancement Techniques
- Multicultural Socio-Legal Studies
- Image and Object Detection Techniques
Southwest Minzu University
2020-2025
State Ethnic Affairs Commission
2024-2025
The First Affiliated Hospital, Sun Yat-sen University
2024
Sun Yat-sen University
2024
Xinjiang Institute of Ecology and Geography
2024
Chinese Academy of Sciences
2024
Vision Technology (United States)
2023
The Engineering & Technical College of Chengdu University of Technology
2013-2020
University of Electronic Science and Technology of China
2017-2020
Chengdu University of Technology
2013-2020
The fragile alpine vegetation in the Tibetan Plateau (TP) is very sensitive to environmental changes, making TP one of hotspots for studying response climate change. Existing studies lack detailed description different climatic factors using method multiple nested time series analysis and grey correlation analysis. In this paper, based on Normalized Difference Vegetation Index (NDVI) growing season calculated from MOD09A1 data product Moderate-resolution Imaging Spectroradiometer (MODIS),...
Available evidence suggests that dynamic functional connectivity can capture time-varying abnormalities in brain activity resting-state cerebral magnetic resonance imaging (rs-fMRI) data and has a natural advantage uncovering mechanisms of abnormal schizophrenia (SZ) patients. Hence, an advanced network analysis model called the temporal category graph convolutional (Temporal-BCGCN) was employed. Firstly, unique module, DSF-BrainNet, designed to construct synchronization features....
Pansharpening is a very debated spatio-spectral fusion problem. It refers to the of high spatial resolution panchromatic image with lower but higher spectral multispectral in order obtain an both domains. In this article, we propose novel variational optimization-based (VO) approach address issue incorporating outcome deep convolutional neural network (DCNN). This solution can take advantages paradigms. On one hand, performance be expected introducing machine learning (ML) methods based on...
The Inner Mongolia Autonomous Region (IMAR) is a major source of rivers, catchment areas, and ecological barriers in the northeast China, related to nation’s security improvement environment. Therefore, studying response vegetation climate change has become an important part current global research. Since existing studies lack detailed descriptions different climatic factors using method grey correlation analysis based on pixel, temporal spatial patterns trends enhanced index (EVI) are...
Nowadays, intelligent medicine is gaining widespread attention, and great progress has been made in Western with the help of artificial intelligence to assist decision making. Compared medicine, traditional Chinese (TCM) involves selecting specific treatment method, prescription, medication based on dialectical results each patient's symptoms. For this reason, development a TCM-assisted decision-making system lagged. Treatment syndrome differentiation core TCM treatment; doctors can...
Cloud removal is an important task in the remotely sensed images (RSIs) processing, which beneficial for downstream applications, such as unmixing, fusion, and target detection. Multi-temporal (MRSIs), contains abundant spatial-spectral-temporal (SST) information, potentially bring new opportunities cloud removal. However, how to effectively efficiently explore rich information of MRSIs remains a challenge. Inspired by low-rankness MRSIs, we propose Unsupervised Domain Factorization Network...
China has abundant grassland resources (approximately 400 million ha of natural grasslands), which account for 41.7% China's total area. Grasslands are an important base boosting the development livestock husbandry economy and maintaining ecological security. Using Moderate Resolution Imaging Spectroradiometer (MODIS) remotely sensed data, this study developed a vegetation growth index that ranked magnitude indices across wide variety field experiments. This applied to conduct remote-sensing...
In recent years, sparse unmixing has attracted significant attention, as it can effectively avoid the bottleneck problems associated with absence of pure pixels and estimation number endmembers in hyperspectral scenes. The joint-sparsity model outperformed single method. However, might cause some aliasing artifacts for on boundaries different constituent endmembers. To address this shortcoming, researchers have developed many algorithms based low-rank representation, which makes good use...
On May 31, 2003, the scan line corrector (SLC) of Enhanced Thematic Mapper Plus (ETM+) on-board Landsat-7 satellite failed, resulting in strips data lost all ETM+ images acquired since then. In this paper, we proposed a novel inpainting algorithm for recovering SLC-off images. The two slopes boundaries each missing stripe were extracted through Hough transform, ignoring slope edge strip that overlaps image. An adaptive dictionary was then developed and trained using SLC-on before 2003 so...
The image signal is represented by using the atomic of to train an over complete dictionary and described as sparse linear combinations these atoms. Recently, algorithm for tracking decomposition mainly adopted focus research. An alternate iterative encoding, sample based on update process K-SVD decomposition. A new segmentation brain MRI image, which uses noise reduction method with adaptive genetic algorithm, presented in this paper, experimental results show that has fast calculation...
In this study, the heating of ions with a Kappa (κ) velocity distribution by an Alfvén wave propagating along external magnetic field via non-resonant wave–particle interactions in low-beta plasmas is investigated means both quasilinear theory and test-particle simulation. The κ most suitable for describing non-thermal enhanced high-energy tail includes Boltzmann as limiting case. Because thermal non-equilibrium factor (κ factor), effect weaker than that ions, when κ→∞, effects become...
In recent years, remote sensing of phenology has become a useful tool to reveal the response and feedback vegetation dynamics global climate change due its multitemporal phase, wide-coverage, continuous space coverage, long-time series. Based on Moderate Resolution Imaging Spectroradiometer data product MOD09A1, enhanced index (EVI) was calculated, time series EVI growing season in Qinghai–Tibet Plateau (QTP) from 2001 2016 reconstructed light double logistic function filter. A dynamic...
Surface reconstruction plays a pivotal role in various fields, including reverse engineering, and oil gas exploration. However, errors available data insufficient surface morphology information often introduce uncertainty into the reconstruction. It is crucial to accurately characterize visualize for risk analysis planning further collection. To this end, paper proposes an characterization method based on twin support vector regression. First, modeling are effectively integrated contained...
Schizophrenia is a serious psychiatric disorder. Its pathogenesis not completely clear, making it difficult to treat patients precisely. Because of the complicated non-Euclidean network structure human brain, learning critical information from brain networks remains difficult. To effectively capture topological neural networks, novel multimodal graph attention based on sparse interaction mechanism (Multi-SIGATnet) was proposed for SZ classification classification. Firstly, structural and...