Proper Feature Space Research Articles

An automated hypersphere-based healthy subspace method for robust and unsupervised damage detection via random vibration response signals

A novel, unsupervised, hypersphere-based healthy subspace method for robust damage detection under non-quantifiable uncertainty via a limited number of random vibration response sensors is postulated. The method is based on the approximate construction, within a proper feature space, of a healthy subspace representing the healthy structural dynamics under uncertainty as the union of properly selected hyperspheres. This is achieved via a fully automated algorithm eliminating user intervention, and thus subjective selections, or complex optimization procedures. The main asset of the proposed method lies in combining simplicity and full automation with high performance. Its performance is systematically assessed via two experimental case studies featuring various uncertainty sources and distinct healthy subspace geometries, while interesting comparisons with three well-known robust damage detection methods are also performed. The results indicate excellent detection performance, which also compares favorably to that of alternative methods.

Convolutional Subspace Clustering Network With Block Diagonal Prior

Standard methods of subspace clustering are based on self-expressiveness in the original data space, which states that a data point in a subspace can be expressed as a linear combination of other points. However, the real data in raw form are usually not well aligned with the linear subspace model. Therefore, it is crucial to obtain a proper feature space for performing high quality subspace clustering. Inspired by the success of Convolutional Neural Networks (CNN) for extraction powerful features from visual data and the block diagonal prior for learning a good affinity matrix from self-expression coefficients, in this paper, we propose a jointly trainable feature extraction and affinity learning framework with the block diagonal prior, termed as Convolutional Subspace Clustering Network with Block Diagonal prior (ConvSCN-BD), in which we solve the joint optimization problem in ConvSCN-BD via an alternating minimization algorithm, which updates the parameters in the convolutional modules and the self-expression coefficients with stochastic gradients descent and updates other variables with close-form solutions alternatingly. In addition, we derive the connection between the block diagonal prior and the subspace structured norm, and reveal that using the block diagonal prior on the affinity matrix is essentially incorporating the feedback information from spectral clustering. Experiments on three benchmark datasets demonstrated the effectiveness of our proposal.

A stochastic Functional Model based method for random vibration based robust fault detection under variable non–measurable operating conditions with application to railway vehicle suspensions

The problem of random vibration based robust fault detection under variable and non–measurable Environmental and Operating Conditions (EOCs) is considered, and a novel stochastic Functional Model (FM) based method is postulated. It is a data–driven method, of the Statistical Time Series (STS) type, and aims at overcoming the well known drawbacks of available methods by achieving high detection performance while eliminating their drawbacks, such as the need for measurable EOCs, for measurement of a high number of vibration signals for proper training, for subjective judgement in selecting method parameters, and for high dimensional non–convex optimization procedures. The method is based on representing the system dynamics, under any set of EOCs, in a proper feature space, within which the healthy dynamics are represented by a proper healthy subspace constructed via a Functional Model. Fault detection is then based upon determining, at a certain risk level, whether or not the current dynamics resides within the healthy subspace. The method's assessment is achieved via simulation results with a case study pertaining to fault detection in a railway vehicle suspension under variable payload, with high detection performance, clearly exceeding that of an alternative Principal Component Analysis (PCA) based method.

Fuzzy Clustering in Cascaded Feature Space

The success of fuzzy clustering heavily relies on the proper feature space constructed by the input data. For nonspherical and overlapped clusters, kernel fuzzy clustering is more effective because it finds more proper feature space compared to conventional fuzzy clustering. Unfortunately, poor scalability of kernel fuzzy clustering is induced by the construction of a kernel matrix. To solve the problem, random feature-based method was presented to approximate the kernel function. More interestingly, these features exposed in the approximate feature space are directly manipulable. Inspired by the architecture of functional-link neural network, to exploit more information from both the original data space and the approximate kernel space, a new feature space called cascaded feature space (CF) is constructed in this paper. By performing classical fuzzy c-means (FCM) in CF space, a new fuzzy clustering framework called FCM-CF is developed. To reduce computational complexity and perform FCM in CF space well, dimension reduction methods are adopted to generate two variants of FCM-CF. The experiment results of our proposed algorithms verify their superiority in comparison of other classical fuzzy clustering algorithms.

Dimensionality Reduction Based on Group-Based Tensor Model for Hyperspectral Image Classification

Dimensionality reduction is one of the most important tasks in improving hyperspectral image (HSI) classification performance and has been widely studied. In traditional dimensionality reduction methods, the hyperspectral data cube is first converted to a matrix, and then, the matrix-algebra is employed, in which the spatial structure is usually not taken into consideration. To jointly utilize spectral and spatial information, researches considering HSI as a tensor have attracted more and more attention. In this letter, we propose a group-based tensor model for HIS dimensionality reduction. The local and nonlocal spatial information of HSI cubes is explored by segmenting the original HSI tensors into a lot of small tensors and grouping them into clusters. Finally, the clusters are projected into low-rank subspace to obtain a proper feature space. Experimental results on two real hyperspectral data sets exhibit the effectiveness of the proposed algorithm for HSI-dimensionality reduction.

A robust perception based method for iris tracking

The paper presents an application of the modified kernel based object tracking to iris tracking. Human perception rules are used for defining a proper feature space for iris that mainly accounts for the fact that eyes represent instinctive visual fixation regions. In addition, a similarity metric that is close to the way human vision compares and perceives the difference between scene components has been employed for finding the exact iris location in subsequent frames. As a result, just one iteration of the mean shift algorithm allows us to get a faithful estimation of iris location in subsequent frames. This property makes the proposed algorithm implementable on mobile devices and useful for real time applications. Experimental results performed on MICHE database show that the proposed method is robust to changes in illumination or scale, iris partial or total occlusion for some subsequent frames, blinks.