位置：成果数据库 > 期刊 > 期刊详情页
稀疏表示保持的鉴别特征选择算法
  • ISSN号：1004-373X
  • 期刊名称：《现代电子技术》
  • 时间：0
  • 分类：TN911-34[电子电信—通信与信息系统;电子电信—信息与通信工程]
  • 作者机构：[1]南京理工大学计算机科学与工程学院,江苏南京210094
  • 相关基金：国家自然科学基金（61202134）; 国家杰出青年科学基金（61125305）; 中国博士后科学基金（AD41431）; 江苏省博士后科学基金
作者： 夏广胜[1], 严慧[1]
关键词： 特征选择, 稀疏表示, 重构残差, l2, 1范数, feature selection, sparse representation, reconstruction residual, 1-norm
中文摘要：

稀疏表示作为一种基于部分数据的表示,已经吸引了越来越多的关注,并广泛应用于模式识别和机器学习领域。提出一种新的算法,称为稀疏表示保持的鉴别特征选择（SRPFS）,其目的是选择鉴别性特征子集,使得在所选特征子空间中,样本的稀疏类内重构残差和稀疏类间重构残差的差值最小化。与传统算法选择特征的独立性方式不同,该算法以批处理方式选择最具鉴别性的特征,并用于优化提出的l2,1范数最小化的目标函数。在标准UCI数据集和哥伦比亚图像数据库的实验结果表明,该算法在识别性能和稳定性方面优于其他经典特征选择算法。

英文摘要：

A new algorithm for selection of distinguishable features preserved by sparse representation,whose aim is to se-lect a subset of distinguishable features to minimize the difference value of reconstruction residual in sparse class and reconstruc-tion residuals between sparse classes of samples in the subspace of selected features. The algorithm,which is different from the selection feature independence mode of the traditional algorithms,selects the most distinguishable features in batch mode and, is used to optimize the minimized objective function of l2,1-norm. The experimental results on standard UCI datasets and Colum-bia object image data base show that the algorithm is superior to other classic feature selection algorithms in the aspects of recog-nition performance and stability.

同期刊论文项目
模式识别与智能系统
期刊论文 91
特征选择中的全局最优搜索策略研究
期刊论文 11 会议论文 8
同项目期刊论文
非负二维主成分分析及在人脸识别中的应用
无监督的差分鉴别特征提取以及在人脸识别上的应用
Sparsity preserving score for joint feature selection
Correntropy based feature selection using binary projection
From classifiers to discriminators: A nearest neighbor rule induced discriminant analysis
Smooth twin support vector regression
Sparse margin-based discriminant analysis for feature extraction
A nonnegative sparse representation based fuzzy similar neighbor classifier
Regularized least squares fisher linear discriminant with applications to image recognition
Sparse Representation Classifier Steered Discriminative Projection With Applications to Face Recogni
Discriminative histograms of local dominant orientation (D-HLDO) for biometric image feature extract
Robustness analysis of static routing on networks
Iterative path attacks on networks
Higher-level feature combination via multiple kernel learning for image classification
Matrix Variate Distribution-Induced Sparse Representation for Robust Image Classification
Bayesian sample steered discriminative regression for biometric image classification
Robust Subspace Segmentation Via Low-Rank Representation
Modified Principal Component Analysis: An Integration of Multiple Similarity Subspace Models
Recursive projection twin support vector machine via within-class variance minimization
Image segmentation by iterated region merging with localized graph cuts
Integration of multiple orientation and texture information for finger-knuckle-print verification
Dimensionality reduction for microarray data using local mean based discriminant analysis
Sparse maximum margin discriminant analysis for feature extraction and gene selection on gene expres
Identifying Subspace Gene Clusters from Microarray Data Using Low-Rank Representation
A flexible support vector machine for regression
Saliency-based localising active contour for automatic natural object segmentation
Enhanced iterative projection for subclass discriminant analysis under EM-alike framework
Reconstruction Based Finger-Knuckle-Print Verification With Score Level Adaptive Binary Fusion
Weighted linear embedding: utilizing local and nonlocal information sufficiently
Regularized Robust Coding for Face Recognition
K-local hyperplane distance nearest neighbor classifier oriented local discriminant analysis
Low-rank representation based discriminative projection for robust feature extraction
Recursive robust least squares support vector regression based on maximum correntropy criterion
Component-based global k-NN classifier for small sample size problems
A novel multiphase active contour model for inhomogeneous image segmentation
Tunable path centrality: Quantifying the importance of paths in networks
Learning image manifold via local tensor subspace alignment
Mixed Noise Removal by Weighted Encoding With Sparse Nonlocal Regularization
A novel sparse representation based framework for face image super-resolution
Continuous attractors of higher-order recurrent neural networks with infinite neurons
Histogram of visual words based on locally adaptive regression kernels descriptors for image feature
Linear reconstruction measure steered nearest neighbor classification framework
Joint Laplacian feature weights learning
Median-mean line based discriminant analysis
Epidemic spreading driven by biased random walks
Locality preserving score for joint feature weights learning
Mixed noise removal by weighted low rank model
Inhomogeneity-embedded active contour for natural image segmentation
Nuclear-L1 norm joint regression for face reconstruction and recognition with mixed noise
Sparse discriminative feature selection
Double Nuclear Norm-Based Matrix Decomposition for Occluded Image Recovery and Background Modeling
Nuclear Norm-Based 2-DPCA for Extracting Features From Images
Robust nuclear norm regularized regression for face recognition with occlusion
Nearest orthogonal matrix representation for face recognition
Optimal Locality Regularized Least Squares Support Vector Machine via Alternating Optimization
Sparse nonlocal priors based two-phase approach for mixed noise removal
Locality-Constrained Sparse Auto-Encoder for Image Classification
Local Structure-Based Image Decomposition for Feature Extraction With Applications to Face Recogniti
Neighbors' distribution property and sample reduction for support vector machines
稀疏表示和贪婪搜索的人脸分类
鉴别的局部中值保持投影及其在人脸识别中的应用
非凸加权核范数及其在运动目标检测中的应用
层次分类方法综述
期刊信息
  • 《现代电子技术》
  • 北大核心期刊（2014版）
  • 主管单位:陕西省信息产业厅
  • 主办单位:陕西电子杂志社 陕西省电子技术研究所
  • 主编：张郁（执行）
  • 地址：西安市金花北路176号陕西省电子技术研究所科研生产大楼六层
  • 邮编：710032
  • 邮箱：met@xddz.com.cn
  • 电话：029-93228979
  • 国际标准刊号：ISSN：1004-373X
  • 国内统一刊号：ISSN：61-1224/TN
  • 邮发代号:52-126
  • 获奖情况:
  • 中国科技核心期刊
  • 国内外数据库收录:
  • 波兰哥白尼索引,中国中国科技核心期刊,中国北大核心期刊（2014版）
  • 被引量:37245