A5103113022- Remote-Sensing Image Classification
- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Remote Sensing and Land Use
- Remote Sensing and LiDAR Applications
- Advanced Image Fusion Techniques
- Advanced X-ray Imaging Techniques
- Soil erosion and sediment transport
- Air Quality and Health Impacts
- RNA and protein synthesis mechanisms
- Energy Load and Power Forecasting
- Digital Holography and Microscopy
- Machine Learning in Bioinformatics
- Genetic Associations and Epidemiology
- Advanced machining processes and optimization
- Evaluation Methods in Various Fields
- Glass properties and applications
- Labor market dynamics and wage inequality
- Data Management and Algorithms
- Livestock and Poultry Management
- Automated Road and Building Extraction
- Ocular and Laser Science Research
- Asthma and respiratory diseases
- X-ray Spectroscopy and Fluorescence Analysis
- Photorefractive and Nonlinear Optics
Chongqing University of Posts and Telecommunications
2020-2025
Shanghai University of Finance and Economics
2025
Southern Medical University
2024
Yangzhou University
2024
Henan Agricultural University
2024
Beijing Normal University
2018-2020
State Key Laboratory of Remote Sensing Science
2018-2020
Wuhan University of Technology
2018
Chinese Academy of Sciences
2011-2018
Institute of Soil Science
2018
Tuberculosis is killing millions of lives every year and on the blacklist most appalling public health problems. Recent findings suggest that secretory protein Mycobacterium tuberculosis may serve purpose developing specific vaccines drugs due to their antigenicity. Responding global infectious disease, we focused identification proteins in . A novel method called MycoSec was designed by incorporating<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"...
Mycobacterium tuberculosis is a bacterium that causes tuberculosis, one of the most prevalent infectious diseases. Predicting subcellular localization mycobacterial proteins in this may provide vital clues for prediction protein function as well drug discovery and design. Therefore, computational method can predict with high precision highly desirable. We propose to proteins. An objective strict benchmark dataset was constructed after collecting 272 non-redundant from universal resource (the...
Multilabel remote sensing (RS) image annotation is a challenging and time-consuming task that requires considerable amount of expert knowledge. Most existing RS methods are based on handcrafted features require multistage processes not sufficiently efficient effective. An can be assigned with single label at the scene level to depict overall understanding multiple labels object represent major components. The used as supervised information for annotation, whereas additional exploit...
Accurate detection of land cover changes on the basis very-high-resolution remote sensing is essential for many practical applications, like urban management, ecological health monitoring, and disaster loss assessment. Modern change (CD) methods are mostly based convolutional neural network (CNN). Despite good performance, CNN-based restricted by receptive field size lack ability exploring long-range dependencies in space–time. To address this issue, we propose a transformer-based CD model,...
Low-rank representation (LRR) can construct the relationships among pixels for hyperspectral image (HSI) classification with a given dictionary and noise term. However, accuracy of HSI based on LRR methods is degraded redundant information existed in pixels. The neglect semantic around may cause "salt-and-pepper" problem classification. To avoid aforementioned problems, novel self-supervised low-rank method called SSLRR developed. In SSLRR, spectral–spatial graph regularization are developed...
The challenges in hyperspectral image (HSI) classification lie the existence of noisy spectral information and lack contextual among pixels. Considering three different levels HSIs, i.e., subpixel, pixel, superpixel, offer complementary information, we develop a novel HSI feature learning network (HSINet) to learn consistent features by self-supervision for classification. HSINet contains three-layer deep neural multifeature convolutional network. It automatically extracts such as spatial,...
The performance of hyperspectral image (HSI) classification relies on the pixel information obtained from hundreds contiguous and narrow spectral bands. Existing approaches, however, are limited to exploit an appropriate latent subspace for data representation within pixel-level or superpixel-level. To utilize spatial correlation among pixels in HSI avoid "salt-and-pepper" problem generated pixel-based classification, a novel superpixel-level aware learning method called PSASL is developed....
Extracting buildings from high-resolution remote sensing images is essential for many geospatial applications, such as building change detection, urban planning, and disaster emergency assessment. Due to the diversity of geometric shapes blurring boundaries among buildings, it still a challenging task accurately generate footprints complex scenes images. The rapid development convolutional neural networks presenting both new opportunities challenges with respect extraction To capture...
Gully erosion can lead to the destruction of farmland and reduction in crop yield. mapping from remote sensing images is critical for quickly obtaining distribution gullies at regional scales arranging corresponding prevention control measures. The narrow irregular shapes similar colors surrounding make sloping challenging. To implement gully mapping, we developed a small training samples-oriented lightweight deep leaning model, called asymmetric non-local LinkNet (ASNL-LinkNet)....
Nitrogen and phosphorus excessive enrichment are major causes of water eutrophication, variations in nutrients strongly influenced by human activities. In this study, annual average quality from 2001 to 2018 was used explore the spatiotemporal total nitrogen (TN) (TP) their relationships with Spatially, TN TP concentrations exhibited significant across five sub-lake zones, values were relatively higher NW lake zone than other zones. Temporally, concentration weak correlations years (R2 =...
Introduction With the rapid development of digital agriculture, digitalization has gradually become a key factor affecting resilience China’s pig farming industry. Methods From both test results and theoretical point view, individual fixed-effect model is more suitable for this paper’s study. Therefore, based on panel data 31 provinces in China from 2011 to 2022, study constructs an examine how digitization affects Results discussion The findings indicated that significantly enhanced hog...
A compact dual-band planar monopole antenna for 2.4/5GHz WLAN application is presented. The two resonant modes of the proposed are associate with four slots which contribute to upper frequency and can reduce size lower length. parametric study performed understand characteristics antenna. Experimental results show that designed provide excellent performance system, including sufficiently wide band, moderate gain, nearly omnidirectional radiation coverage.
It has been shown that grating based x‐ray phase contrast imaging (GPI) can be performed on tubes and therefore a great potential for wide application in many fields. So far to extract separate the information from other contributions, phase‐stepping approach is normally adopted. One of gratings scanned transversely incident beam while acquiring multiple projections sample. However, during scanning, sample supposed static, resulting poor time resolution one major drawbacks this method. clear...
Band selection (BS) work for hyperspectral remote sensing images (HRSIs) has attracted increasing attention from researchers. In the HRSI, salient features of each band always include more important discriminative features. However, most existing BS methods have difficulty in simultaneously performing local-global consistency analysis on bands HRSIs. addition, these also ignore consideration diversity information to some extent when selecting a subset. To tackle problems, novel unsupervised...
Abstract Snakehead vesiculovirus (SHVV) is one of the primary pathogens responsible for viral diseases in snakehead fish. A TaqMan-based real-time PCR assay was established rapid detection and quantification SHVV this study. Specific primers fluorescent probes were designed phosphoprotein (P) gene, after optimizing reaction conditions, results indicated that limit method could reach 37.1 copies, representing a 100-fold increase sensitivity compared to RT-PCR. The specificity testing revealed...