•An autoencoder based framework for structural damage identification is proposed.•It supports deep neural networks for solution of pattern recognition problems.•Dimensionality reduction and ...relationship learning are included in the framework.•Accurate identification results are obtained considering uncertainties and noises.•Numerical and experimental investigations are conducted to validate the approach.
Artificial neural networks are computational approaches based on machine learning to learn and make predictions based on data, and have been applied successfully in diverse applications including structural health monitoring in civil engineering. It is difficult to optimize the weights in the neural networks that have multiple hidden layers due to the vanishing gradient issue. This paper proposes an autoencoder based framework for structural damage identification, which can support deep neural networks and be utilized to obtain optimal solutions for pattern recognition problems of highly non-linear nature, such as learning a mapping between the vibration characteristics and structural damage. Two main components are defined in the proposed framework, namely, dimensionality reduction and relationship learning. The first component is to reduce the dimensionality of the original input vector while preserving the required necessary information, and the second component is to perform the relationship learning between the features with the reduced dimensionality and the stiffness reduction parameters of the structure. Vibration characteristics, such as natural frequencies and mode shapes, are used as the input and the structural damage are considered as the output vector. A pre-training scheme is performed to train the hidden layers in the autoencoders layer by layer, and fine tuning is conducted to optimize the whole network. Numerical and experimental investigations on steel frame structures are conducted to demonstrate the accuracy and efficiency of the proposed framework, comparing with the traditional ANN methods.
Palmprint is one important biometric feature with uniqueness, stability and high distinguishability, and its study has attracted much attention in the past decades. Although many palmprint-based ...recognition methods have been proposed and successfully applied for identity authentication, most of the previous researches usually only use the images captured in natural light. It is hard, if not impossible, for further improvement of recognition accuracy based on these palmprint images due to limitations of using the natural light. In order to obtain high recognition rate with more discriminative information, we propose to use multispectral palmprint instead of natural light palmprint in this paper, and develop a multispectral palmprint recognition method based on a hierarchical idea. First, we extract the Block Dominant Orientation Code (BDOC) as a rough feature, and the Block-based Histogram of Oriented Gradient (BHOG) as a fine feature. Second, a hierarchical recognition approach is proposed based on these two types of features. Technically, we fuse different features obtained from different bands in the proposed scheme in order to improve the recognition accuracy. Finally, experimental results show that the recognition accuracy of the proposed method is not only superior to previous high-performance methods based on the PolyU palmprint database with the natural light but also it can further improve the state of the art performance achieved by some approaches based on the PolyU multispectral palmprint database.
•The characteristics of the multispectral palmprint image are more distinctive than the natural light image.•A hierarchical idea is used to improve the recognition accuracy.•The BDOC method and BHOG method are proposed to extract the robust orientation features from the multispectral palmprint image.•To further improve the recognition accuracy, a feature level fusion scheme based on the hierarchical idea is used to integrate the multispectral information.
Given
n training examples, the training of a least squares support vector machine (LS-SVM) or kernel ridge regression (KRR) corresponds to solving a linear system of dimension
n. In cross-validating ...LS-SVM or KRR, the training examples are split into two distinct subsets for a number of times
(
l
)
wherein a subset of
m
examples are used for validation and the other subset of
(
n
-
m
)
examples are used for training the classifier. In this case
l
linear systems of dimension
(
n
-
m
)
need to be solved. We propose a novel method for cross-validation (CV) of LS-SVM or KRR in which instead of solving
l
linear systems of dimension
(
n
-
m
)
, we compute the inverse of an
n dimensional square matrix and solve
l
linear systems of dimension
m, thereby reducing the complexity when
l
is large and/or
m is small. Typical multi-fold, leave-one-out cross-validation (LOO-CV) and leave-many-out cross-validations are considered. For five-fold CV used in practice with five repetitions over randomly drawn slices, the proposed algorithm is approximately four times as efficient as the naive implementation. For large data sets, we propose to evaluate the CV approximately by applying the well-known incomplete Cholesky decomposition technique and the complexity of these approximate algorithms will scale linearly on the data size if the rank of the associated kernel matrix is much smaller than
n
. Simulations are provided to demonstrate the performance of LS-SVM and the efficiency of the proposed algorithm with comparisons to the naive and some existent implementations of multi-fold and LOO-CV.
This paper investigates a granular description for facial expression intensity in which characterization of motion-based expression features is presented by a collection of information granules. ...First, the motion-based features are extracted based on the facial landmarks to encode the expression intensity. Secondly, the semantic concepts are then generated by involving various mechanisms of fuzzy clustering based on the motion-based features. The resulting clusters can be viewed as numeric prototypes of the descriptors. The information granules are formed around the prototypes according to the principle of justifiable granularity, which is the fundamental idea of Granular Computing. The proposed granular descriptor for expression intensity is tested on the BU-3DFE database, and the experimental results illustrate that the proposed descriptors based on information granules not only can characterize the semantics of facial changes caused by expressions, but also can facilitate the semantic estimation of facial expression intensity.
Underwater captured images often suffer from poor visibility caused by two major degradations: scattering and absorption. In this paper, we propose a hybrid framework for underwater image ...enhancement, which unifies underwater white balance and variational contrast and saturation enhancement. In our framework, the improved underwater white balance (UWB) algorithm is integrated with histogram stretching, aiming to better compensate the attenuation difference along the propagation path and remove undesired color castings. In addition, a variational contrast and saturation enhancement (VCSE) model is developed based on the enhanced result obtained from UWB. The advantages of VCSE model lie in the improvements of contrast and saturation as well as the elimination of hazy appearance induced by scattering. Moreover, we design a fast Gaussian pyramid-based algorithm to speed up the solving of VCSE model. The improvements achieved by our method include the more effective in color correction, haze removal and detail clarification. Extensive qualitative and quantitative assessments demonstrate that the proposed approach obtains high quality outcomes, which outperforms several state-of-the-art methods. Application tests further verify the effectiveness and broad application prospects of our proposed method.
In this paper, we propose a new learning algorithm, named as the Cooperative and Geometric Learning Algorithm (CGLA), to solve problems of maneuverability, collision avoidance and information sharing ...in path planning for Unmanned Aerial Vehicles (UAVs). The contributions of CGLA are three folds: (1) CGLA is designed for path planning based on cooperation of multiple UAVs. Technically, CGLA exploits a new defined individual cost matrix, which leads to an efficient path planning algorithm for multiple UAVs. (2) The convergence of the proposed algorithm for calculating the cost matrix is proven theoretically, and the optimal path in terms of path length and risk measure from a starting point to a target point can be calculated in polynomial time. (3) In CGLA, the proposed individual weight matrix can be efficiently calculated and adaptively updated based on the geometric distance and risk information shared among UAVs. Finally, risk evaluation is introduced first time in this paper for UAV navigation and extensive computer simulation results validate the effectiveness and feasibility of CGLA for safe navigation of multiple UAVs.
► We investigate the drawbacks for clustering algorithm DBSCAN. ► We use the affinity propagation algorithm to detect the density of a dataset. ► We propose a new free parameter clustering algorithm ...APSCAN. ► APSCAN is validated via extensive experiments.
DBSCAN is a density based clustering algorithm and its effectiveness for spatial datasets has been demonstrated in the existing literature. However, there are two distinct drawbacks for DBSCAN: (i) the performances of clustering depend on two specified parameters. One is the maximum radius of a neighborhood and the other is the minimum number of the data points contained in such neighborhood. In fact these two specified parameters define a single density. Nevertheless, without enough prior knowledge, these two parameters are difficult to be determined; (ii) with these two parameters for a single density, DBSCAN does not perform well to datasets with varying densities. The above two issues bring some difficulties in applications. To address these two problems in a systematic way, in this paper we propose a novel parameter free clustering algorithm named as APSCAN. Firstly, we utilize the Affinity Propagation (AP) algorithm to detect local densities for a dataset and generate a normalized density list. Secondly, we combine the first pair of density parameters with any other pair of density parameters in the normalized density list as input parameters for a proposed DDBSCAN (Double-Density-Based SCAN) to produce a set of clustering results. In this way, we can obtain different clustering results with varying density parameters derived from the normalized density list. Thirdly, we develop an updated rule for the results obtained by implementing the DDBSCAN with different input parameters and then synthesize these clustering results into a final result. The proposed APSCAN has two advantages: first it does not need to predefine the two parameters as required in DBSCAN and second, it not only can cluster datasets with varying densities but also preserve the nonlinear data structure for such datasets.
The classical non-local Total Variation model has been extensively used for gray texture image inpainting previously, but such model can not be directly applied to color texture image inpainting due ...to coupling of different image channels in color images. In order to solve the inpainting problem for color texture images effectively, we propose a non-local Color Total Variation model. This model is different from the recently proposed non-local Mumford–Shah model (NL-MS). Technically, the proposed model is an extension of local TV model for gray images but we take account of the relationship between different channels in color images and make use of concepts of the non-local operators. We will analyze how the coupling of different channels of color images in the proposed model makes the problem difficult for numerical implementation with the conventional split Bregman algorithm. In order to solve the proposed model efficiently, we propose a fast heuristic numerical algorithm based on the split Bregman algorithm with introduction of a threshold function. The performance of the proposed model with the proposed heuristic algorithm is compared with the NL-MS model. Extensive numerical experiments have shown that the proposed model and algorithm have superior excellent performance as well as with much faster speed.
Euler’s elastica-based unsupervised segmentation models have strong capability of completing the missing boundaries for existing objects in a clean image, but they are not working well for noisy ...images. This paper aims to establish a Euler’s elastica-based approach that can properly deal with the random noises to improve the segmentation performance for noisy images. The corresponding formulation of stochastic optimization is solved via the progressive hedging algorithm (PHA), and the description of each individual scenario is obtained by the alternating direction method of multipliers. Technically, all the sub-problems derived from the framework of PHA can be solved by using the curvature-weighted approach and the convex relaxation method. Then, an alternating optimization strategy is applied by using some powerful accelerating techniques including the fast Fourier transform and generalized soft threshold formulas. Extensive experiments have been conducted on both synthetic and real images, which displayed significant gains of the proposed segmentation models and demonstrated the advantages of the developed algorithms.