A5101993208- Geophysical Methods and Applications
- interferon and immune responses
- Microwave Imaging and Scattering Analysis
- Image Retrieval and Classification Techniques
- MicroRNA in disease regulation
- Remote-Sensing Image Classification
- RNA modifications and cancer
- Ultrasonics and Acoustic Wave Propagation
- Non-Destructive Testing Techniques
- Advanced Neural Network Applications
- Antibiotic Resistance in Bacteria
- Remote Sensing and LiDAR Applications
- Fluid Dynamics and Turbulent Flows
- Heat Transfer Mechanisms
- Remote Sensing and Land Use
- Aquaculture disease management and microbiota
- Domain Adaptation and Few-Shot Learning
- Maritime Navigation and Safety
- Machine Learning and ELM
- Ubiquitin and proteasome pathways
- Image Processing Techniques and Applications
- Military Strategy and Technology
- Advanced Algorithms and Applications
- Advanced Computational Techniques and Applications
- Military Defense Systems Analysis
Guangdong Polytechnic Normal University
2024
Guangxi University
2024
Shandong University
2023-2024
Weifang Medical University
2022-2023
Changsha University
2022
Dali University
2019
Northwestern Polytechnical University
2017-2018
Beihang University
2018
Beijing Radiation Center
2012-2013
National University of Defense Technology
2013
More recently, remote sensing image classification has been moving from pixel-level interpretation to scene-level semantic understanding, which aims label each scene with a specific class. While significant efforts have made in developing various methods for classification, most of them rely on handcrafted features. In this letter, we propose novel feature representation method named bag convolutional features (BoCF). Different the traditional visual words-based words are usually obtained by...
Hyperspectral image (HSI) classification is an active and important research task driven by many practical applications. To leverage deep learning models especially convolutional neural networks (CNNs) for HSI classification, this paper proposes a simple yet effective method to extract hierarchical spatial feature exploring the power of off-the-shelf CNN models, without any additional retraining or fine-tuning on target data set. obtain better accuracy, we further propose unified metric...
The influenza virus (IV) triggers a series of signalling events inside host cells and induces complex cellular responses. Studies have suggested that factors play an essential role in IV replication. MicroRNAs (miRNAs) represent class small non-coding RNAs target mRNAs, triggering either translation repression or RNA degradation. Emerging research suggests host-derived miRNAs are involved mediating the host-IV interaction. Using miRNA microarrays, we identified several aberrantly expressed...
In this paper, we propose a feature transformation ensemble model with batch spectral regularization for the Cross-domain few-shot learning (CD-FSL) challenge. Specifically, proposes to construct an prediction by performing diverse transformations after extraction network. On each branch network of use term suppress singular values matrix during pre-training improve generalization ability model. The proposed can then be fine tuned in target domain address classification. We also further...
A deep-neural-network-based identification method for multipolarimetric ground-penetrating radar (GPR) data was proposed to address the challenges of unbalanced polarimetric and take advantage GPR subsurface defect identification. The comprised two subnetworks—namely, an augmentation module based on a cycle generative adversarial network (CycleGAN) generate large number paired data, accurately identify internal defects with irregular shapes occlusions. Specifically, physics-informed...
Abstract Periodic inspections and recognition of temporal evolution for concrete internal defects are important the long-term operation civil engineering infrastructures. This study proposes a method integrating convolutional neural network (CNN) with maximum intensity projection (MIP) 3D imaging defects. The proposed utilizes MIP to process multiple ground penetrating radar (GPR) B-scans generates 2D projected GPR containing spatial information. SegNet coupled Lovász softmax loss function...
<title>Abstract</title> Underwater object detection (UOD) is crucial for marine ecosystem protection, ocean resource exploration, and maritime security. While generic methods have shown success in atmospheric conditions, they fall short the underwater environment. The complicated environment physical properties of light make it highly challenging to distinguish ambiguous objects from background. Also, common phenomenon wild that small aquatic creatures often congregate large groups further...
Human miRNAs have recently been found to important roles in viral replication. Understanding the patterns and details of human miRNA interactions during virus-host may help uncover novel antiviral therapies. Based on abundance knowledge available regarding protein-protein (PPI), protein interactions, experimentally validated miRNA-target pairs transcriptional regulation miRNAs, it is possible explore complex regulatory network that exists between proteins at system level.By integrating...
The opportunistic pathogen Pseudomonas aeruginosa often adapts to its host environment and causes recurrent nosocomial infections. extracytoplasmic function (ECF) sigma factor enables bacteria alter their gene expression in response environmental stimuli. Here, we report an ECF factor, HxuI, which is rapidly induced once P. encounters the host. Host stresses such as iron limitation, oxidative stress, low oxygen, nitric oxide induce of hxuI. By combining RNA-seq promoter-lacZ reporter fusion...
Abstract With long-term operation of tunnels, the surface tunnel lining is prone to crack, spalling, water leakage and other defects. Therefore, detection defects has become an important process ensure safety operation. Linear scan cameras have main way defect because their ultrahigh resolution scanning frequency. However, in collecting images at high speed, linear are easy cause distortion, which leads reduction identification accuracy. Thus, this paper proposes a distortion image...
In order to predict the epidemic outbreak, this paper presented a wavelet neural network method. The designing theory framework was also provided: function is used as activation in recessive layer, connection weights are decided by coefficients. Finally, method efficiency and feasibility were discussed simulation example, some key problems research directions summarized..
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI
Abstract This paper studied the process of distributed methane purification and proposed a method to collect separate carbon dioxide from by simulation. Aspen Hysys software was used simulate effect hydrogen sulfide content change on compression, separation, decarbonization process. The critical properties methane, power consumption compressor, caloric value were qualitatively quantitatively analyzed. Therefore, this article has significant guidance how improve high value-added utilization...
Mycobacterium tuberculosis (Mtb) is the bacterial pathogen responsible for causing (TB), a severe public health concern that results in numerous deaths worldwide. Ubiquitination (Ub) an essential physiological process aids maintaining homeostasis and contributes to development of TB. Therefore, main objective our study was investigate potential role Ub-related genes TB.Our research entailed utilizing single sample gene set enrichment analysis (ssGSEA) combination with several machine...
Spiders are the most abundant predatory natural enemies in terrestrial ecosystems. As an important enemy of many agricultural and forestry pests, spiders play a very significant role biological control pests. In order to make rational use spider resources, it is necessary observe study population characteristics it. Direct observation method time-consuming laborious. If we can take videos or images by surveillance cameras, then computer vision technology identify classify automatically,...
Traditional game theory for modeling strategies of attacker and defender are static or discrete, such models do not adequately address real world situation in which the attackers' investments dynamic, so defenders'. By contrast, risk analysis generally neglects fact that terrorists intelligent adaptive, can observe our defenses adapt their accordingly. Considering both aspects circumstances, with motivation reflect adaptive nature terrorist threats structures systems to be defended, this...
The signals of reinforcing steel bars (rebars) in ground penetrating radar (GPR) images may mask defect underneath them, negatively affecting detection reinforced concrete. This study proposes an unsupervised learning-based method trained using unpaired GPR to suppress the rebar signal and reconstruct on recorded B-Scan images. In particular, four contrastive feature encoders two similarity are designed network. extracted features constrained by loss functions, respectively. proposed was...