自适应嵌入的半监督多视角特征降维方法

半监督模式下的多视角特征降维方法,大多并未考虑到不同视角间特征投影的差异,且由于缺乏对降维后的低维矩阵的稀疏约束,无法避免噪声和其他不相关特征的影响。针对这两个问题,提出自适应嵌入的半监督多视角特征降维方法。首先,将投影从单视角下相同的嵌入矩阵扩展到多视角间不同的矩阵,引入全局结构保持项;然后,将无标签的数据利用无监督方法进行嵌入投影,对于有标签的数据,结合分类的判别信息进行线性投影;最后,再将两类多投影映射到统一的低维空间,使用组合权重矩阵来保留全局结构,很大程度上消除了噪声及不相关因素的影响。实验结果表明,所提方法的聚类准确率平均提高了约9%。该方法较好地保留了多视角间特征的相关性,捕获了更多的具有判别信息的特征。     

Locality preserving embedding for face and handwriting digital recognition

Most supervised manifold learning-based methods preserve the original neighbor relationships to pursue the discriminating power. Thus, structure information of the data distributions might be neglected and destroyed in low-dimensional space in a certain sense. In this paper, a novel supervised method, called locality preserving embedding (LPE), is proposed to feature extraction and dimensionality reduction. LPE can give a low-dimensional embedding for discriminative multi-class sub-manifolds and preserves principal structure information of the local sub-manifolds. In LPE framework, supervised and unsupervised ideas are combined together to learn the optimal discriminant projections. On the one hand, the class information is taken into account to characterize the compactness of local sub-manifolds and the separability of different sub-manifolds. On the other hand, at the same time, all the samples in the local neighborhood are used to characterize the original data distributions and preserve the structure in low-dimensional subspace. The most significant difference from existing methods is that LPE takes the distribution directions of local neighbor data into account and preserves them in low-dimensional subspace instead of only preserving the each local sub-manifold’s original neighbor relationships. Therefore, LPE optimally preserves both the local sub-manifold’s original neighborhood relationships and the distribution direction of local neighbor data to separate different sub-manifolds as far as possible. The criterion, similar to the classical Fisher criterion, is a Rayleigh quotient in form, and the optimal linear projections are obtained by solving a generalized Eigen equation. Furthermore, the framework can be directly used in semi-supervised learning, and the semi-supervised LPE and semi-supervised kernel LPE are given. The proposed LPE is applied to face recognition (on the ORL and Yale face databases) and handwriting digital recognition (on the USPS database). The experimental results show that LPE consistently outperforms classical linear methods, e.g., principal component analysis and linear discriminant analysis, and the recent manifold learning-based methods, e.g., marginal Fisher analysis and constrained maximum variance mapping.     

无监督混阶栈式稀疏自编码器的图像分类学习

目前多数图像分类的方法是采用监督学习或者半监督学习对图像进行降维,然而监督学习与半监督学习需要图像携带标签信息。针对无标签图像的降维及分类问题,提出采用混阶栈式稀疏自编码器对图像进行无监督降维来实现图像的分类学习。首先,构建一个具有三个隐藏层的串行栈式自编码器网络,对栈式自编码器的每一个隐藏层单独训练,将前一个隐藏层的输出作为后一个隐藏层的输入,对图像数据进行特征提取并实现对数据的降维。其次,将训练好的栈式自编码器的第一个隐藏层和第二个隐藏层的特征进行拼接融合,形成一个包含混阶特征的矩阵。最后,使用支持向量机对降维后的图像特征进行分类,并进行精度评价。在公开的四个图像数据集上将所提方法与七个对比算法进行对比实验,实验结果表明,所提方法能够对无标签图像进行特征提取,实现图像分类学习,减少分类时间,提高图像的分类精度。     

图Laplacian和自训练用于高光谱数据半监督波段选择

波段选择是数据降维的有效手段,但有限的标记样本影响了监督波段选择的性能。提出一种利用图Laplacian和自训练策略实现半监督波段选择的方法。该方法首先定义基于图的半监督特征评分准则以产生初始波段子集,接着在该子集基础上进行分类,采用自训练策略将部分可信度较高的非标记样本扩展至标记样本集合,再用特征评分准则对波段子集进行更新。重复该过程,获得最终波段子集。高光谱波段选择与分类实验比较了多种非监督、监督和半监督方法,实验结果表明所提算法能选择出更好的波段子集。     

Unsupervised spectral sub-feature learning for hyperspectral image classification

Spectral pixel classification is one of the principal techniques used in hyperspectral image (HSI) analysis. In this article, we propose an unsupervised feature learning method for classification of hyperspectral images. The proposed method learns a dictionary of sub-feature basis representations from the spectral domain, which allows effective use of the correlated spectral data. The learned dictionary is then used in encoding convolutional samples from the hyperspectral input pixels to an expanded but sparse feature space. Expanded hyperspectral feature representations enable linear separation between object classes present in an image. To evaluate the proposed method, we performed experiments on several commonly used HSI data sets acquired at different locations and by different sensors. Our experimental results show that the proposed method outperforms other pixel-wise classification methods that make use of unsupervised feature extraction approaches. Additionally, even though our approach does not use any prior knowledge, or labelled training data to learn features, it yields either advantageous, or comparable, results in terms of classification accuracy with respect to recent semi-supervised methods.     

SSPS: A Semi-Supervised Pattern Shift for Classification

Recently, a great amount of efforts have been spent in the research of unsupervised and (semi-)supervised dimensionality reduction (DR) techniques, and DR as a preprocessor is widely applied into classification learning in practice. However, on the one hand, many DR approaches cannot necessarily lead to a better classification performance. On the other hand, DR often suffers from the problem of estimation of retained dimensionality for real-world data. Alternatively, in this paper, we propose a new semi-supervised data preprocessing technique, named semi-supervised pattern shift (SSPS). The advantages of SSPS lie in the fact that not only the estimation of retained dimensionality can be avoided naturally, but a new shifted pattern representation that may be more favorable to classification is obtained as well. As a further extension of SSPS, we develop its fast and out-of-sample versions respectively, both of which are based on a shape-preserved subset selection trick. The final experimental results demonstrate that the proposed SSPS is promising and effective in classification application.     

基于二进制生成对抗网络的视觉回环检测研究

针对现有的回环检测模型大多基于有监督学习进行训练,需要大量标注数据的问题,提出一种视觉回环检测新方法,利用生成对抗思想设计一个深度网络,以无监督学习的方式训练该网络并提取高区分度和低维度的二进制特征。将距离传播损失函数和二值化表示熵损失函数引入神经网络,将高维特征空间的海明距离关系传播到低维特征空间并增加低维特征表示的多样性,进而利用BoVW模型将提取的局部特征融合为全局特征用于回环检测。实验结果表明:相比SIFT和ORB等特征提取方法,所述方法在具有强烈视角变化和外观变化的复杂场景下具有更好的性能,可以与AlexNet和AMOSNet等有监督深度网络相媲美。但采用无监督学习,从根本上避免了费时费力的数据标注过程,特别适用于大规模开放场景的回环检测,同时二进制特征描述符极大地节约了存储空间和计算资源。     

Discriminative deep belief networks for visual data classification

Visual data classification using insufficient labeled data is a well-known hard problem. Semi-supervise learning, which attempts to exploit the unlabeled data in additional to the labeled ones, has attracted much attention in recent years. This paper proposes a novel semi-supervised classifier called discriminative deep belief networks (DDBN). DDBN utilizes a new deep architecture to integrate the abstraction ability of deep belief nets (DBN) and discriminative ability of backpropagation strategy. For unsupervised learning, DDBN inherits the advantage of DBN, which preserves the information well from high-dimensional features space to low-dimensional embedding. For supervised learning, through a well designed objective function, the backpropagation strategy directly optimizes the classification results in training dataset by refining the parameter space. Moreover, we apply DDBN to visual data classification task and observe an important fact that the learning ability of deep architecture is seriously underrated in real-world applications, especially in visual data analysis. The comparative experiments on standard datasets of different types and different scales demonstrate that the proposed algorithm outperforms both representative semi-supervised classifiers and existing deep learning techniques. For visual dataset, we can further improve the DDBN performance with much larger and deeper architecture.     

Joint self-representation and subspace learning for unsupervised feature selection

This paper proposes a novel unsupervised feature selection method by jointing self-representation and subspace learning. In this method, we adopt the idea of self-representation and use all the features to represent each feature. A Frobenius norm regularization is used for feature selection since it can overcome the over-fitting problem. The Locality Preserving Projection (LPP) is used as a regularization term as it can maintain the local adjacent relations between data when performing feature space transformation. Further, a low-rank constraint is also introduced to find the effective low-dimensional structures of the data, which can reduce the redundancy. Experimental results on real-world datasets verify that the proposed method can select the most discriminative features and outperform the state-of-the-art unsupervised feature selection methods in terms of classification accuracy, standard deviation, and coefficient of variation.     

监督式谱空间分类器

提出了一种非线性的监督式谱空间分类器(supervised spectral space classifier,简称S3C).S3C首先将输入数据映射到融合了训练数据判别信息的低维监督式谱空间中,然后在该监督式谱空间中构造最大化间隔的最优分割超平面,并把测试数据以无监督的方式也映射到与训练数据相同的新特征空间中,最后,直接应用之前构建的分类超平面对映射后的测试数据进行分类.由于S3C使研究者可以直观地观察到变化后的特征空间和映射后的数据,因此有利于对算法的评价和参数的选择.在S3C的基础上,进一步提出了一种监督式谱空间分类器的改进算法(supervised spectral space transformation,简称S3T).S3T通过采用线性子空间变换和强迫一致的方法,将映射到监督式谱空间内的数据再变换到指定的类别指示空间中去,从而获得关于测试数据的类别指示矩阵,并在此基础上对其进行分类.S3T不仅保留了S3C算法的各项优点,而且还可以用于直接处理多分类问题,抗噪声能力更强,性能更加鲁棒.在人工数据集和真实数据集上的大量实验结果显示,S3C和S3T与其他多种著名分类器相比,具有更加优越的分类性能.     

Spatial-spectral locality preserving projection for hyperspectral image classification with limited training samples

Locality preserving projection (LPP) is a popular unsupervised feature extraction (FE) method. In this paper, the spatial-spectral LPP (SSLPP) method is proposed, which uses both the spectral and spatial information of hyperspectral image (HSI) for FE. The proposed method consists of two parts. In the first part, unlabelled samples are selected in a spatially homogeneous neighbourhood from filtered HSI. In the second part, the transformation matrix is calculated by an LPP-based method and by using the spectral and spatial information of the selected unlabelled samples. Experimental results on Indian Pines (IP), Kennedy Space Center (KSC), and Pavia University (PU) datasets show that the performance of SSLPP is superior to spectral unsupervised, supervised, and semi-supervised FE methods in small and large sample size situations. Moreover, the proposed method outperforms other spatial-spectral semi-supervised FE methods for PU dataset, which has high spatial resolution. For IP and KSC datasets, spectral regularized local discriminant embedding (SSRLDE) has the best performance by using spectral and spatial information of labelled and unlabelled samples, and SSLPP is ranked just behind it. Experiments show that SSLPP is an efficient unsupervised FE method, which does not use training samples as preparation of them is so difficult, costly, and sometimes impractical. SSLPP results are much better than LPP. Also, it decreases the storage and calculation costs using less number of unlabelled samples.     

一种新的基于特殊离群样本优化的三维点云特征 选择算法

随着元宇宙、数字孪生、虚拟现实与增强现实等前沿技术的快速发展,三维点云在电力、建筑、 先进制造等行业中得到广泛应用,随之而来的,如何降低三维点云数据冗余度、有效进行点云特征选择,已在充 分利用海量点云数据中扮演着关键角色。考虑到现有大多数三维点云特征选择算法忽略了特定样本在特征评估中 的表现,提出一种新的有监督特征选择算法,即基于特殊离群样本优化的特征选择算法(FSSO)。具体地,为获得 精准的特殊离群样本(SOs),FSSO 优化均值中心并动态地界定类簇主体;计算 SOs 的类内相对偏离程度,通过减 小类内相对偏离对特征进行打分,实现特征选择过程。在 3 个公共的三维点云模型分类数据集上(ModelNet40, IntrA,ShapeNetCore)的实验,以及 4 个高维人工特征数据集的验证实验结果表明,相较于其他特征选择算法, FSSO 可选择出具有更强分类能力的特征子集,并提升分类准确率。     

Brain image segmentation using semi-supervised clustering

The objective of brain image segmentation is to partition the brain images into different non-overlapping homogeneous regions representing the different anatomical structures. Magnetic resonance brain image segmentation has large number of applications in diagnosis of neurological disorders like Alzheimer diseases, Parkinson related syndrome etc. But automatically segmenting the MR brain image is not an easy task. To solve this problem, several unsupervised and supervised based classification techniques have been developed in the literature. But supervised classification techniques are more time consuming and cost-sensitive due to the requirement of sufficient labeled data. In contrast, unsupervised classification techniques work without using any prior information but it suffers from the local trap problems. So, to overcome the problems associated with unsupervised and supervised classification techniques, we have proposed a new semi-supervised clustering technique using the concepts of multiobjective optimization and applied this technique for automatic segmentation of MR brain images in the intensity space. Multiple centers are used to encode a cluster in the form of a string. The proposed clustering technique utilizes intensity values of the brain pixels as the features. Additionally it also assumes that the actual class label information of 10% points of a particular image data set is also known. Three cluster validity indices are utilized as the objective functions, which are simultaneously optimized using AMOSA, a modern multiobjective optimization technique based on the concepts of simulated annealing. First two cluster validity indices are symmetry distance based Sym-index and Euclidean distance based I-index, which are based on unsupervised properties. Last one is a supervised information based cluster validity index, Minkowski Index. The effectiveness of this proposed semi-supervised clustering technique is demonstrated on several simulated MR normal brain images and MR brain images having some multiple sclerosis lesions. The performance of the proposed semi-supervised clustering technique is compared with some other popular image segmentation techniques like Fuzzy C-means, Expectation Maximization and some recent image clustering techniques like multi-objective based MCMOClust technique, and Fuzzy-VGAPS clustering techniques.     

面向社会事件的半监督自训练多方立场分析

已有的立场分析方法主要采用有监督或无监督方式训练立场分类模型,有监督模型训练通常需要大量有标注数据支持,而相比有监督模型,无监督模型的性能差距较大.为了降低模型训练对有标注训练数据的要求,同时保证模型性能,文中面向社会事件相关的社交媒体文本,提出半监督自训练多方立场分析方法.对于自训练方法,在模型迭代训练过程中,选择高质量样本加入训练集合,对提升模型性能起到关键作用.为此,文中方法首先根据用户立场一致性度量文本的分类置信度,然后利用话题信息进一步筛选高质量样本扩充训练集合,保证模型性能不断提升.实验表明,相比相关工作中的代表性方法和其它半监督模型训练方式,文中方法能够取得更优的立场分类效果,并且方法依据的用户立场一致性和话题信息均有助于提升立场分类效果.     

Application of Kernel-Based Feature Space Transformations and Learning Methods to Phoneme Classification

This paper examines the applicability of some learning techniques to the classification of phonemes. The methods tested were artificial neural nets (ANN), support vector machines (SVM) and Gaussian mixture modeling (GMM). We compare these methods with a traditional hidden Markov phoneme model (HMM), working with the linear prediction-based cepstral coefficient features (LPCC). We also tried to combine the learners with linear/nonlinear and unsupervised/supervised feature space transformation methods such as principal component analysis (PCA), independent component analysis (ICA), linear discriminant analysis (LDA), springy discriminant analysis (SDA) and their nonlinear kernel-based counterparts. We found that the discriminative learners can attain the efficiency of HMM, and that after the transformations they can retain the same performance in spite of the severe dimension reduction. The kernel-based transformations brought only marginal improvements compared to their linear counterparts.     

Unsupervised feature selection via joint local learning and group sparse regression

Feature selection has attracted a great deal of interest over the past decades. By selecting meaningful feature subsets, the performance of learning algorithms can be effectively improved. Because label information is expensive to obtain, unsupervised feature selection methods are more widely used than the supervised ones. The key to unsupervised feature selection is to find features that effectively reflect the underlying data distribution. However, due to the inevitable redundancies and noise in a dataset, the intrinsic data distribution is not best revealed when using all features. To address this issue, we propose a novel unsupervised feature selection algorithm via joint local learning and group sparse regression (JLLGSR). JLLGSR incorporates local learning based clustering with group sparsity regularized regression in a single formulation, and seeks features that respect both the manifold structure and group sparse structure in the data space. An iterative optimization method is developed in which the weights finally converge on the important features and the selected features are able to improve the clustering results. Experiments on multiple real-world datasets (images, voices, and web pages) demonstrate the effectiveness of JLLGSR.

     

Semi-supervised dimension reduction based on hypergraph embedding for hyperspectral images

Dimension reduction (DR) is an efficient and effective preprocessing step of hyperspectral images (HSIs) classification. Graph embedding is a frequently used model for DR, which preserves some geometric or statistical properties of original data set. The embedding using simple graph only considers the relationship between two data points, while in real-world application, the complex relationship between several data points is more important. To overcome this problem, we present a linear semi-supervised DR method based on hypergraph embedding (SHGE) which is an improvement of semi-supervised graph learning (SEGL). The proposed SHGE method aims to find a projection matrix through building a semi-supervised hypergraph which can preserve the complex relationship of the data and the class discrimination for DR. Experimental results demonstrate that our method achieves better performance than some existing DR methods for HSIs classification and is time saving compared with the existed method SEGL which used simple graph.     

半监督学习研究的述评

监督学习需要利用大量的标记样本训练模型,但实际应用中,标记样本的采集费时费力。无监督学习不使用先验信息,但模型准确性难以保证。半监督学习突破了传统方法只考虑一种样本类型的局限,能够挖掘大量无标签数据隐藏的信息,辅助少量的标记样本进行训练,成为机器学习的研究热点。通过对半监督学习研究的总趋势以及具体研究内容进行详细的梳理与总结,分别从半监督聚类、分类、回归与降维以及非平衡数据分类和减少噪声数据共六个方面进行综述,发现半监督方法众多,但存在以下不足：（1）部分新提出的方法虽然有效,但仅通过特定数据集进行了实证,缺少一定的理论证明;（2）复杂数据下构建的半监督模型参数较多,结果不稳定且缺乏参数选取的指导经验;（3）监督信息多采用样本标签或成对约束形式,对混合约束的半监督学习需要进一步研究;（4）对半监督回归的研究匮乏,对如何利用连续变量的监督信息研究甚少。     

Deep and fast: Deep learning hashing with semi-supervised graph construction

Learning-based hashing methods are becoming the mainstream for approximate scalable multimedia retrieval. They consist of two main components: hash codes learning for training data and hash functions learning for new data points. Tremendous efforts have been devoted to designing novel methods for these two components, i.e., supervised and unsupervised methods for learning hash codes, and different models for inferring hashing functions. However, there is little work integrating supervised and unsupervised hash codes learning into a single framework. Moreover, the hash function learning component is usually based on hand-crafted visual features extracted from the training images. The performance of a content-based image retrieval system crucially depends on the feature representation and such hand-crafted visual features may degrade the accuracy of the hash functions. In this paper, we propose a semi-supervised deep learning hashing (DLH) method for fast multimedia retrieval. More specifically, in the first component, we utilize both visual and label information to learn an optimal similarity graph that can more precisely encode the relationship among training data, and then generate the hash codes based on the graph. In the second stage, we apply a deep convolutional network to simultaneously learn a good multimedia representation and a set of hash functions. Extensive experiments on five popular datasets demonstrate the superiority of our DLH over both supervised and unsupervised hashing methods.     

极化合成孔径雷达图像相干斑抑制和分类方法综述

极化合成孔径雷达（Polarimetric synthetic aperture radar,PolSAR）是一种多参数、多通道的微波成像系统,在农林业、地质、海洋和军事等领域有着广泛的应用前景.PolSAR图像的相干斑抑制和分类是数据解译的重要环节,已经成为遥感领域的研究热点.本文综述了现有PolSAR图像的相干斑噪声抑制和分类方法并进行展望.首先,简要介绍了PolSAR系统的主要进展和应用;然后,对PolSAR图像相干斑抑制的评价指标和方法进行综述并对几种代表性方法进行了实验对比;接下来,对PolSAR图像的特征进行分析归纳,分别对有监督、无监督和半监督的PolSAR分类方法进行总结并给出了几种有监督分类方法的实验比较;最后,对PolSAR图像相干斑抑制和分类方法未来可能的研究方向进行了思考和讨论.