A5062567062- Face and Expression Recognition
- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Gene expression and cancer classification
- Machine Learning and Algorithms
- Machine Learning and Data Classification
- Medical Image Segmentation Techniques
- Sparse and Compressive Sensing Techniques
- Neural Networks and Applications
- Statistical Methods and Inference
- Distributed Sensor Networks and Detection Algorithms
- Machine Learning in Bioinformatics
- Advanced Statistical Methods and Models
- Cognitive Radio Networks and Spectrum Sensing
- Wireless Communication Security Techniques
- Advanced Algorithms and Applications
- Statistical Distribution Estimation and Applications
- Bayesian Modeling and Causal Inference
- Williams Syndrome Research
- Machine Learning and ELM
- Probability and Risk Models
- Control Systems and Identification
- Image Processing and 3D Reconstruction
- Algorithms and Data Compression
- Spectroscopy and Chemometric Analyses
Missouri State University
2014-2023
Ludong University
2023
York University
2019
University of California, Los Angeles
2006-2011
Xi'an Jiaotong University
2003-2004
Efficient and reliable spectrum sensing plays a critical role in cognitive radio networks. This paper presents cooperative sequential detection scheme to reduce the average time that is required reach decision. In scheme, each computes log-likelihood ratio for its every measurement, base station sequentially accumulates these statistics determines whether stop making measurement. The studies how implement robust manner when assumed signal models have unknown parameters, such as strength...
The decision tree method has grown fast in the past two decades and its performance classification is promising. tree-based ensemble algorithms have been used to improve of an individual tree. In this study, we compared four basic methods, that is, bagging tree, random forest, AdaBoost terms size, band selection (BS), feature selection, accuracy efficiency ecological zone Clark County, Nevada, through multi-temporal multi-source remote-sensing data. Furthermore, BS schemes based on...
It is important to detect and extract the major cortical sulci from brain images, but manually annotating these a time-consuming task requires labeler follow complex protocols. This paper proposes learning-based algorithm for automated extraction of magnetic resonance imaging (MRI) volumes surfaces. Unlike alternative methods detecting sulci, which use small number predefined rules based on properties surface such as mean curvature, our approach learns discriminative model using...
Object boundary detection and segmentation is a central problem in computer vision. The importance of combining low-level, mid-level, high-level cues has been realized recent literature. However, it unclear how to efficiently effectively engage fuse different levels information. In this paper, we emphasize learning based approach explore information, both implicitly explicitly. First, learn low-level for object boundaries interior regions using probabilistic boosting tree (PBT). Second,...
Efficient and reliable spectrum sensing plays a critical role in cognitive radio networks. This paper presents cooperative sequential detection scheme to minimize the average time that is required reach decision. In scheme, each computes Log-Likelihood ratio for its every measurement, base station sequentially accumulates these statistics determines whether stop making measurement. The number of samples depends on Kullback-Leibler distance between distributions two hypotheses under test....
This paper describes a coarse-to-fine learning based image registration algorithm which has particular advantages in dealing with multi-modality images. Many existing algorithms [18] use few designed terms or mutual information to measure the similarity between pairs. Instead, we push aspect by selecting and fusing large number of features for measuring similarity. Moreover, is carried strategy: global first performed roughly locate component, then learn/compute on local patches capture fine...
High-spatial-resolution aerial photographs can provide detailed distribution of sea ice features. However, very few studies have ever considered shadows on the for detection. In this letter, shadows, retrieved from 163 selected acquired July 26, 2010, in a marginal-ice-zone area near Barrow, Alaska, utilizing an object-based classification scheme, are used to estimate ridge attributes through local solar illumination geometry. The photograph-averaged frequency (354.6-8908.7 km <sup...
To overcome the vast computation of standard support vector machines (SVMs), Lee and Mangasarian (see First SIAM International Conference on Data Mining, 2001) proposed reduced (RSVM). But they select 'support vectors' randomly from training set, this will affect test result. In paper, we some representative vectors as via a simple unsupervised clustering algorithm, then apply RSVM method these vectors. The can get higher recognition accuracy with fewer compared to original RSVM, advantage...
This paper gives an improvement to Bennett's inequality for tail probability of sum independent random variables, without imposing any additional condition. The improved version has a closed form expression. Using refined arithmetic-geometric mean inequality, we further improve the obtained inequality. Numerical comparisons show that proposed inequalities often upper bound significantly in far area, and these improvements get more prominent larger sample size.
Efficient and reliable spectrum sensing plays a critical role in cognitive radio networks. This paper proposes cooperative scheme that detects the existence of common signal component signals received by multiple geographically distributed radios. The assumes different radios display strong coherence if they have source. Detection this wireless environment is studied, especially when transmitted distorted multipath channels.
Nonnegative matrix factorization is used extensively for feature extraction and clustering analysis. Recently many sparsity/sparseness constraints, such as L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> penalty, are introduced sparse nonnegative factorization. Inspired by sparsity measures from linear regression model, this paper proposes to integrate with another constraint, the elastic net. The experimental results of analysis on...
This paper presents a new local edge-based level set model that does not use initial contours. Unlike traditional active contours gradient to detect edges, our derives the neighborhood distribution and edge information with two different localized region-based operators: Gaussian mixture model-based intensity estimator Hueckel operator. We incorporate operator outcomes into recently proposed binary fitting (LBF) as (LDF) model, which enables without contour selection, i.e., function can be...