A5015739024- Image and Signal Denoising Methods
- Medical Image Segmentation Techniques
- Stability and Controllability of Differential Equations
- Advanced Image Fusion Techniques
- Advanced Image Processing Techniques
- Advanced Neural Network Applications
- Nonlinear Dynamics and Pattern Formation
- Remote-Sensing Image Classification
- Advanced Mathematical Physics Problems
- Advanced Image and Video Retrieval Techniques
- Magnesium Alloys: Properties and Applications
- Navier-Stokes equation solutions
- Advanced Mathematical Modeling in Engineering
- Aluminum Alloys Composites Properties
- Image Processing Techniques and Applications
- Video Surveillance and Tracking Methods
- Spectroscopy and Chemometric Analyses
- Advanced Malware Detection Techniques
- Fractional Differential Equations Solutions
- Colorectal Cancer Surgical Treatments
- Network Security and Intrusion Detection
- Colorectal and Anal Carcinomas
- Orthopaedic implants and arthroplasty
- Nonlinear Partial Differential Equations
- Sparse and Compressive Sensing Techniques
Hunan University of Technology
2023-2024
Beijing Normal University
2009-2024
First Affiliated Hospital of Hebei Medical University
2024
Hebei Medical University
2024
First Automotive Works (China)
2022
Northwestern Polytechnical University
2022
Nanchang Hangkong University
2019-2021
Huaqiao University
2018-2019
Zhejiang University
2018
Ministry of Education of the People's Republic of China
2010
This paper proposes a general weighted <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">l</i> <sup xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> - xmlns:xlink="http://www.w3.org/1999/xlink">0</sup> norms energy minimization model to remove mixed noise such as Gaussian-Gaussian mixture, impulse noise, and Gaussian-impulse from the images. The approach is built upon maximum likelihood estimation framework sparse representations over trained...
Non-invasive, real-time imaging and deep focus into tissue are in high demand biomedical research. However, the aberration that is introduced by refractive index inhomogeneity of biological hinders way forward. A rapid focusing with sensor-less corrections, based on machine learning, demonstrated this paper. The proposed method applies Convolutional Neural Network (CNN), which can rapidly calculate low-order aberrations from point spread function images Zernike modes after training. results...
In this paper, the traditional model based variational methods and deep learning algorithms are naturally integrated to address mixed noise removal, specially for Gaussian mixture Gaussian-impulse removal problem. To be different from single type (e.g. Gaussian) it is a challenge problem accurately discriminate types levels each pixel. We propose method iteratively estimate parameters, then algorithm can automatically classify according statistical parameters. The proposed separated into...
In this paper, we propose an image segmentation model that incorporates convexity shape priori using level set representations. the past decade, several discrete and continuous methods have been developed to solve problem. Our method comes from observation signed distance function of a convex region must be function. Based on observation, transfer complicated geometrical into some simple constraints We algorithm keep these exactly. The proposed could easily applied based models, such as...
Hyperspectral image (HSI) super-resolution refers to enhancing the spatial resolution of a 3-D with many spectral bands (slices). It is seriously ill-posed problem when low-resolution (LR) HSI only input. better solved by fusing LR high-resolution (HR) multispectral (MSI) for both high and resolution. In this article, we propose novel nonnegative nonlocal 4-D tensor dictionary learning-based model using group-block sparsity. By grouping similar cubes into clusters then conduct cluster...
To achieve high scene classification performance of spatial resolution remote sensing images (HSR-RSIs), it is important to learn a discriminative space in which the distance metric can precisely measure both similarity and dissimilarity features labels between images. While traditional learning methods focus on preserving interclass separability, label consistency (LC) less involved, this might degrade accuracy. Aiming at considering intraclass compactness HSR-RSIs, we propose method with...
Application-layer distributed denial-of-service (DDoS) attacks incapacitate systems by using up their resources, causing service interruptions, financial losses, and more. Consequently, advanced deep-learning techniques are used to detect mitigate these in cloud infrastructures. However, mobile edge computing (MEC), it becomes economically impractical equip each node with defensive as resources may largely remain unused devices. Furthermore, current methods mainly concentrated on improving...
Vehicle data, which may have some errors, provide a source for big data analysis. Car owners sometimes publish false information to protect their privacy or interests. The spread of these messages will contribute potential loss as rumors. Recently, the emergence vehicular social network has further accelerated rumors and anti-rumors. Researchers proposed number methods reduce caused by However, do not account fact that it takes time users reply message after receiving Each user responds...
The performance of scene classification relies heavily on the spatial and structural features that are extracted from high resolution remote-sensing images. Existing approaches, however, limited in adequately exploiting latent relationships between Aiming to decrease distances intraclass images increase interclass images, we propose a relationship learning framework integrates an adaptive graph with constraints feature space label propagation for high-resolution aerial image classification....
Abstract This article presents the method of computer automatic recognition and measurement number volume nanoparticles formed on a rough surface by smoothing, enhancement segmentation image processing. The grafted grains (nanoparticles) polyethylene are taken as example. uses shock filter globally convex to separate from polymer substrate surface. Then extracted surface, determined. By applying this analyze surfaces irradiated for different time, obtained they consistent with results...
A dual expectation-maximization (EM) algorithm for total variation (TV) regularized Gaussian mixture model (GMM) is proposed in this paper. The built upon the EM with TV regularization (EM-TV) which combines statistical and variational methods together image segmentation. Inspired by projection Chambolle, we give a EM-TV model. related problem smooth can be easily solved gradient method, stable fast. Given parameters of GMM, seen as forward-backward splitting method converges. This extended...
Image segmentation is a fundamental research topic in image processing and computer vision. In recent decades, researchers developed large number of algorithms for various applications. Among these algorithms, the normalized cut (Ncut) method widely applied due to its good performance. The Ncut model an optimization problem whose energy defined on specifically designed graph. Thus, results existing are largely dependent preconstructed similarity measure graph since this usually given...
Mixture-of-Experts (MoE) models have gained popularity in achieving state-of-the-art performance a wide range of tasks computer vision and natural language processing. They effectively expand the model capacity while incurring minimal increase computation cost during training. However, deploying such for inference is difficult due to their large size complex communication pattern. In this work, we provide characterization two MoE workloads, namely Language Modeling (LM) Machine Translation...