A5101522875- Wireless Signal Modulation Classification
- Machine Fault Diagnosis Techniques
- Advanced Algorithms and Applications
- Advanced Measurement and Detection Methods
- Neural Networks and Applications
- Advanced Computational Techniques and Applications
- Blind Source Separation Techniques
- Railway Engineering and Dynamics
- Anomaly Detection Techniques and Applications
- Image and Signal Denoising Methods
- Advanced Decision-Making Techniques
- Fault Detection and Control Systems
- Video Surveillance and Tracking Methods
- Evaluation Methods in Various Fields
- Vehicle License Plate Recognition
- Rough Sets and Fuzzy Logic
- Geophysical Methods and Applications
- Optical Systems and Laser Technology
- Advanced Sensor and Control Systems
- Radar Systems and Signal Processing
- Advanced Image Processing Techniques
- Remote Sensing and Land Use
- Network Traffic and Congestion Control
- Network Security and Intrusion Detection
- Infrastructure Maintenance and Monitoring
Southwest Hospital
2022-2025
Army Medical University
2022-2025
Southwest Jiaotong University
2015-2024
Guangxi Transportation Research Institute
2022-2024
Vision Technology (United States)
2023
Nanning Normal University
2020-2022
Shenyang Agricultural University
2020
Wuxi Third People's Hospital
2018
Shandong University of Finance and Economics
2008-2017
Sichuan Normal University
2012
We propose a new multiscale rotation-invariant convolutional neural network (MRCNN) model for classifying various lung tissue types on high-resolution computed tomography. MRCNN employs Gabor-local binary pattern that introduces good property in image analysis-invariance to scales and rotations. In addition, we offer an approach deal with the problems caused by imbalanced number of samples between different classes most existing works, accomplished changing overlapping size adjacent patches....
This paper presents a novel modular recurrent neural network based on the pipelined architecture (PRWNN) to reduce computational complexity and improve performance of wavelet (RWNN). The PRWNN inherits architectures proposed by Haykin Li is made up number RWNN modules that are interconnected in chained form. Since those can be simultaneously performed parallelism fashion, this would lead crucial improvement efficiency. Furthermore, owing cascade interconnection dynamic modules, further...
Exosomal microRNAs (miRNAs) are among the most common biomarkers for tumor diagnosis. However, single-miRNA detection lacks ideal sensitivity and specificity diagnosing a certain in clinics. In this work, we fabricated convenient multi-miRNA platform sensitive specific on exosomal miRNAs plasma of patients using terahertz (THz) metamaterial biosensor basis strand displacement amplification (SDA) AuNPs. The proposed was highly to miRNA-15, miRNA-21, miRNA-145, miRNA-155, miRNA-423, miRNA-451,...
The health management of railway vehicles is crucial to secure safety and efficiency in the long-term operation high-speed trains. Meanwhile, complex components put forward a higher requirement for robustness condition monitoring systems, especially abilities identify unexpected faults. misidentification infrequent faults could lead unpredictable consequences vehicle's safety. This paper proposes novel method detecting train bogie based on Bayesian deep learning. First, Monte Carlo-Based...
Feature extraction is one of key steps in fault diagnosis for High Speed Train (HST). In this work, we present a method that can automatically extract high-level features from HST vibration signals and recognize the faults. The composed Deep Belief Network (DBN) on Fast Fourier Transform (FFT) signals. DBNs be trained greedily, layer by layer, using model referred to as Restricted Boltzmann Machine (RBM). real data sets simulation are selected experiments. First, preprocessed FFT. Then, FFT...
The high sensitivity and specificity of terahertz (THz) biosensing are both promising challenging in DNA sample detection. This study produced refined a flexible THz MM biosensor for ultrasensitive detection HBV clinical serum samples based on gold magnetic nanoparticle-mediated rolling circle amplification (GMNPs@RCA) sandwich assay under isothermal conditions. Typically, solid-phase RCA reactions mediated by circular padlock probes (PLPs) triggered conditions the presence DNA, resulting...
With the development of advanced human-machine interface, effective information processing algorithms, and more powerful microprocessors, it is possible to enable blind achieve additional perception environment. In 1970s, blind-assistant facility based on sensory substitution was introduced. It called as ETA (short for Electronic Travel Aids), bases other natural senses blind, such hearing, touch, smell, feeling etc. Then research vision aids has been broadly extended. After introducing...
A new fire monitoring method was proposed in this paper, which a Neural Network of Deep Convolutional Long-Recurrent Networks (DCLRN), and combining DCLRN network optical flow for open space environment real time. This is achieved by utilizing the static dynamic characteristics fire, converting RGB images to real-time, use convolutional neural spatial learning, class recurrent architectures sequence eventually achieve purpose monitoring. Which end-to-end trainable suitable large-scale visual...
Deep Belief Network (DBN) learns the features of raw data automatically, and develops a new idea for study fault analysis High Speed Train (HST). Combining deep learning classification ensemble technology, this paper presents novel DBN hierarchical model HST analysis. Firstly, Fast Fourier Transform (FFT) coefficients vibration signals are extracted as state visible layer model, then is used to learn automatically. The each learned by train Support Vector Machine (SVM), K-Nearest Neighbor...
Both proper animal welfare and economic benefit are important to the broiler industry, so it is better consider these two factors together. The purpose of this study was investigate relationship between in different production systems white-feathered broilers China. Based on Welfare Quality Assessment (WQA) protocol for poultry, authors compared evaluated results model (WQM) deterministic model. present conducted evaluations investigations 66 chicken flocks 52 farms These included three...
In this paper, we present a new wavelet shrinkage operator to remove noise for non-linear time series. Like other improvements, the proposed is used address discontinuity of hard as well large bias soft operator. The special that it introduces universal threshold divide details signal and noise. Such can identify those coefficients with noise, then shrink larger them, otherwise, small. We test denoising performance in chaotic series generated by Lorenz system. results show method...
In everchanging threat emitter environment, specific identification (SEI) technology extracts subtle but persistent features from received pulse signal to create a fingerprint unique radar. Unlike conventional five parameters deinterleaving algorithm, which can be grossly ambiguous for radar sorting, the SEI provides hardware identification. this paper, we propose an approach extracting unintentional phase modulation caused by oscillator based on surrounding-line integral bispectrum. The...
Image denoising is a challenging task that essential in numerous computer vision and image processing problems. This study proposes applies generative adversarial network-based training architecture to multiple-level Gaussian tasks. Convolutional neural approaches come across blurriness issue produces denoised images blurry on texture details. To resolve the issue, we first performed theoretical of cause problem. Subsequently, proposed an denoiser network, which uses learning process for...
Median filter was once the most popular nonlinear for removing impulse noise because of its good denoising power. In this paper, we present a novel adaptive fuzzy median to restoration salt & pepper noise-corrupted image, which is particularly effective at highly impulsive noise. First estimate level based on fussy set theory, then process corrupted pixel or extend size filtering window, last get appropriate value replace noisy pixel. The proposed has benefits that it simple and assumes no...
Feature-based (FB) algorithms for automatic modulation recognition of radar signals have received much attention since they are usually simple to realize. However, existing FB approaches focus on several specific modulations and fail when applied various modulations. To overcome this issue, we propose a effective algorithm based Manhattan distance-based features (MDBFs) in paper. MDBFs new that can be different The main contributions paper as follows. First, represented wavelet ridges, which...