Results 1  10
of
22
Spectral clustering ensemble applied to sar image segmentation
 MS (w) GBIS (x) Consensus
"... Abstract—Spectral clustering (SC) has been used with success in the field of computer vision for data clustering. In this paper, a new algorithm named SC ensemble (SCE) is proposed for the segmentation of synthetic aperture radar (SAR) images. The graylevel cooccurrence matrixbased statistic featu ..."
Abstract

Cited by 14 (0 self)
 Add to MetaCart
(Show Context)
Abstract—Spectral clustering (SC) has been used with success in the field of computer vision for data clustering. In this paper, a new algorithm named SC ensemble (SCE) is proposed for the segmentation of synthetic aperture radar (SAR) images. The graylevel cooccurrence matrixbased statistic features and the energy features from the undecimated wavelet decomposition extracted for each pixel being the input, our algorithm performs segmentation by combining multiple SC results as opposed to using outcomes of a single clustering process in the existing literature. The random subspace, random scaling parameter, and Nyström approximation for component SC are applied to construct the SCE. This technique provides necessary diversity as well as high quality of component learners for an efficient ensemble. It also overcomes the shortcomings faced by the SC, such as the selection of scaling parameter, and the instability resulted from the Nyström approximation method in image segmentation. Experimental results show that the proposed method is effective for SAR image segmentation and insensitive to the scaling parameter. Index Terms—Image segmentation, spectral clustering (SC), synthetic aperture radar (SAR), unsupervised ensemble. I.
Manifold Clustering of Shapes
 In Proc. of ICDM
, 2006
"... Shape clustering can significantly facilitate the automatic labeling of objects present in image collections. For example, it could outline the existing groups of pathological cells in a bank of cytoimages; the groups of species on photographs collected from certain aerials; or the groups of object ..."
Abstract

Cited by 11 (2 self)
 Add to MetaCart
(Show Context)
Shape clustering can significantly facilitate the automatic labeling of objects present in image collections. For example, it could outline the existing groups of pathological cells in a bank of cytoimages; the groups of species on photographs collected from certain aerials; or the groups of objects observed on surveillance scenes from an office building. Here we demonstrate that a nonlinear projection algorithm such as Isomap can attract together shapes of similar objects, suggesting the existence of isometry between the shape space and a low dimensional nonlinear embedding. Whenever there is a relatively small amount of noise in the data, the projection forms compact, convex clusters that can easily be learned by a subsequent partitioning scheme. We further propose a modification of the Isomap projection based on the concept of degreebounded minimum spanning trees. The proposed approach is demonstrated to move apart bridged clusters and to alleviate the effect of noise in the data. 1.
Graph Transduction as a Noncooperative Game
, 2012
"... Graph transduction is a popular class of semisupervised learning techniques that aims to estimate a classification function defined over a graph of labeled and unlabeled data points. The general idea is to propagate the provided label information to unlabeled nodes in a consistent way. In contrast t ..."
Abstract

Cited by 7 (2 self)
 Add to MetaCart
(Show Context)
Graph transduction is a popular class of semisupervised learning techniques that aims to estimate a classification function defined over a graph of labeled and unlabeled data points. The general idea is to propagate the provided label information to unlabeled nodes in a consistent way. In contrast to the traditional view, in which the process of label propagation is defined as a graph Laplacian regularization, this article proposes a radically different perspective, based on gametheoretic notions. Within the proposed framework, the transduction problem is formulated in terms of a noncooperative multiplayer game whereby equilibria correspond to consistent labelings of the data. An attractive feature of this formulation is that it is inherently a multiclass approach and imposes no constraint whatsoever on the structure of the pairwise similarity matrix, being able to naturally deal with asymmetric and negative similarities alike. Experiments on a number of realworld problems demonstrate that the proposed approach performs well compared with stateoftheart algorithms, and it can deal effectively with various types of similarity relations.
Initialization independent clustering with actively selftraining method
 IEEE Transactions on Systems, Man, and CyberneticsPart B: Cybernetics
, 2012
"... Abstract—The results of traditional clustering methods are usually unreliable as there is not any guidance from the data labels, while the class labels can be predicted more reliable by the semisupervised learning if the labels of partial data are given. In this paper, we propose an actively selftr ..."
Abstract

Cited by 6 (5 self)
 Add to MetaCart
(Show Context)
Abstract—The results of traditional clustering methods are usually unreliable as there is not any guidance from the data labels, while the class labels can be predicted more reliable by the semisupervised learning if the labels of partial data are given. In this paper, we propose an actively selftraining clustering method, in which the samples are actively selected as training set to minimize an estimated Bayes error, and then explore semisupervised learning to perform clustering. Traditional graphbased semisupervised learning methods are not convenient to estimate the Bayes error; we develop a specific regularization framework on graph to perform semisupervised learning, in which the Bayes error can be effectively estimated. In addition, the proposed clustering algorithm can be readily applied in a semisupervised setting with partial class labels. Experimental results on toy data and realworld data sets demonstrate the effectiveness of the proposed clustering method on the unsupervised and the semisupervised setting. It is worthy noting that the proposed clustering method is free of initialization, while traditional clustering methods are usually dependent on initialization. Index Terms—Active learning, initialization independent clustering, selftraining, spectral clustering (SC). I.
Global Linear Neighborhoods for Efficient Label Propagation
"... Graphbased semisupervised learning improves classification by combining labeled and unlabeled data through label propagation. It was shown that the sparse representation of graph by weighted local neighbors provides a better similarity measure between data points for label propagation. However, se ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
(Show Context)
Graphbased semisupervised learning improves classification by combining labeled and unlabeled data through label propagation. It was shown that the sparse representation of graph by weighted local neighbors provides a better similarity measure between data points for label propagation. However, selecting local neighbors can lead to disjoint components and incorrect neighbors in graph, and thus, fail to capture the underlying global structure. In this paper, we propose to learn a nonnegative lowrank graph to capture global linear neighborhoods, under the assumption that each data point can be linearly reconstructed from weighted combinations of its direct neighbors and reachable indirect neighbors. The global linear neighborhoods utilize information from both direct and indirect neighbors to preserve the global cluster structures, while the lowrank property retains a compressed representation of the graph. An efficient algorithm based on a multiplicative update rule is designed to learn a nonnegative lowrank factorization matrix minimizing the neighborhood reconstruction error. Large scale simulations and experiments on UCI datasets and highdimensional gene expression datasets showed that label propagation based on global linear neighborhoods captures the global cluster structures better and achieved more accurate classification results. 1
Semisupervised kernel matrix learning by kernel propagation
 IEEE TNN
, 2010
"... Abstract — The goal of semisupervised kernel matrix learning (SSKML) is to learn a kernel matrix on all the given samples on which just a little supervised information, such as class label or pairwise constraint, is provided. Despite extensive research, the performance of SSKML still leaves some s ..."
Abstract

Cited by 4 (2 self)
 Add to MetaCart
(Show Context)
Abstract — The goal of semisupervised kernel matrix learning (SSKML) is to learn a kernel matrix on all the given samples on which just a little supervised information, such as class label or pairwise constraint, is provided. Despite extensive research, the performance of SSKML still leaves some space for improvement in terms of effectiveness and efficiency. For example, a recent pairwise constraints propagation (PCP) algorithm has formulated SSKML into a semidefinite programming (SDP) problem, but its computation is very expensive, which undoubtedly restricts PCPs scalability in practice. In this paper, a novel algorithm, called kernel propagation (KP), is proposed to improve the comprehensive performance in SSKML. The main idea of KP is first to learn a smallsized subkernel matrix (named seedkernel matrix) and then propagate it into a largersized fullkernel matrix. Specifically, the implementation of KP consists of three stages: 1) separate the supervised sample (sub)set Xl from the full sample set X; 2) learn a seedkernel matrix on Xl through solving a smallscale SDP problem; and 3) propagate the learnt seedkernel matrix into a fullkernel matrix on X. Furthermore, following the idea in KP, we naturally develop two conveniently realizable outofsample extensions for KML: one is batchstyle extension, and the other is onlinestyle extension. The experiments demonstrate that KP is encouraging in both effectiveness and efficiency compared with three stateoftheart algorithms and its related outofsample extensions are promising too. Index Terms — Kernel propagation, outofsample extension, pairwise constraint, seedkernel matrix learning, semidefinite programming. I.
Manifold Learning by Graduated Optimization
"... Abstract—We present an algorithm for manifold learning called manifold sculpting, which utilizes graduated optimization to seek an accurate manifold embedding. An empirical analysis across a wide range of manifold problems indicates that manifold sculpting yields more accurate results than a number ..."
Abstract

Cited by 3 (2 self)
 Add to MetaCart
(Show Context)
Abstract—We present an algorithm for manifold learning called manifold sculpting, which utilizes graduated optimization to seek an accurate manifold embedding. An empirical analysis across a wide range of manifold problems indicates that manifold sculpting yields more accurate results than a number of existing algorithms, including Isomap, locally linear embedding (LLE), Hessian LLE (HLLE), and landmark maximum variance unfolding (LMVU), and is significantly more efficient than HLLE and LMVU. Manifold sculpting also has the ability to benefit from prior knowledge about expected results. Index Terms—Manifold learning, nonlinear dimensionality reduction, unsupervised learning. I.
Fast graph construction using auction algorithm
 In Proceedings of the 28th Annual Conference on Uncertainty in Artificial Intelligence
"... In practical machine learning systems, graph based data representation has been widely used in different learning paradigms, ranging from unsupervised clustering to supervised classification. Besides those applications with natural graph or network structure data, such as social network analysis and ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
(Show Context)
In practical machine learning systems, graph based data representation has been widely used in different learning paradigms, ranging from unsupervised clustering to supervised classification. Besides those applications with natural graph or network structure data, such as social network analysis and relational learning, many other applications often involve a critical step of the conversion from data vectors to a data graph. In particular, a sparse subgraph extracted from the original adjacency graph is required due to both theoretic and practical needs. Previous study clearly shows that the performance of different learning algorithms, e.g., clustering and classification, benefit from such sparse subgraphs with balanced node connectivity. However, the existing graph construction methods are either computationally expensive or with unsatisfactory performance. In this paper, we introduce a scalable method called auction algorithm and its parallel extension to recover a sparse yet balanced subgraph with significantly reduced computational cost. Empirical study and comparison with the stateofart approaches clearly demonstrate the superiority of the proposed method in both efficiency and accuracy. 1
Automatic outlier detection: A Bayesian approach
 In IEEE International Conference on Robotics and Automation
, 2007
"... Abstract — In order to achieve reliable autonomous control in advanced robotic systems like entertainment robots, assistive robots, humanoid robots and autonomous vehicles, sensory data needs to be absolutely reliable, or some measure of reliability must be available. Bayesian statistics can offer f ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
(Show Context)
Abstract — In order to achieve reliable autonomous control in advanced robotic systems like entertainment robots, assistive robots, humanoid robots and autonomous vehicles, sensory data needs to be absolutely reliable, or some measure of reliability must be available. Bayesian statistics can offer favorable ways of accomplishing such robust sensory data preprocessing. In this paper, we introduce a Bayesian way of dealing with outlierinfested sensory data and develop a “black box ” approach to removing outliers in realtime and expressing confidence in the estimated data. We develop our approach in the framework of Bayesian linear regression with heteroscedastic noise. Essentially, every measured data point is assumed to have its individual variance, and the final estimate is achieved by a weighted regression over observed data. An ExpectationMaximization algorithm allows us to estimate the variance of each data point in an incremental algorithm. With the exception of a time horizon (window size) over which the estimation process is averaged, no open parameters need to be tuned, and no special assumption about the generative structure of the data is required. The algorithm works efficiently in realtime. We evaluate our method on synthetic data and on a pose estimation problem of a quadruped robot, demonstrating its ease of usability, competitive nature with welltuned alternative algorithms and advantages in terms of robust outlier removal. I.