Yiguang Liu

The Reliability of Travel Time Forecasting

Travel time is a fundamental measure in transportation, and accurate travel time forecasting is crucial in intelligent transportation systems (ITSs). Currently, many techniques have been applied to travel time forecasting; however, the reliability of the prediction has not been studied in these approaches. In this paper, we propose an approach using the generalized autoregressive conditional heteroscedasticity (GARCH) model to study the volatility of travel time and supply the information about reliability for travel time forecasting. Three examples on real urban vehicular traffic data show the whole modeling process. In the experiments, we utilize the conditional predicted standard deviation (PSD) to express the reliability of travel time forecasting and screen out the sample points that are thought to be reliable forecasting. The results show that the root-mean-square error (RMSE), mean absolute error (MAE), and mean absolute percent error (MAPE) are all decreasing with an increase in the demand of the reliability. It proves that the model well depicts the reliability of travel time forecasting and that the proposed approach is feasible.

A robust face and ear based multimodal biometric system using sparse representation

If fusion rules cannot adapt to the changes of environment and individual users, multimodal systems may perform worse than unimodal systems when one or more modalities encounter data degeneration. This paper develops a robust face and ear based multimodal biometric system using Sparse Representation (SR), which integrates the face and ear at feature level, and can effectively adjust the fusion rule based on reliability difference between the modalities. We first propose a novel index called Sparse Coding Error Ratio (SCER) to measure the reliability difference between face and ear query samples. Then, SCER is utilized to develop an adaptive feature weighting scheme for dynamically reducing the negative effect of the less reliable modality. In multimodal classification phase, SR-based classification techniques are employed, i.e., Sparse Representation based Classification (SRC) and Robust Sparse Coding (RSC). Finally, we derive a category of SR-based multimodal recognition methods, including Multimodal SRC with feature Weighting (MSRCW) and Multimodal RSC with feature Weighting (MRSCW). Experimental results demonstrate that: (a) MSRCW and MRSCW perform significantly better than the unimodal recognition using either face or ear alone, as well as the known multimodal methods; (b) The effectiveness of adaptive feature weighting is verified. MSRCW and MRSCW are very robust to the image degeneration occurring to one of the modalities. Even when face (ear) query sample suffers from 100% random pixel corruption, they can still get the performance close to the ear (face) unimodal recognition; (c) By integrating the advantages of adaptive feature weighting and sparsity-constrained regression, MRSCW seems excellent in tackling the face and ear based multimodal recognition problem.

On the Almost Periodic Solution of Cellular Neural Networks With Distributed Delays

By exponential dichotomy about differential equations, a formal almost periodic solution (APS) of a class of cellular neural networks (CNNs) with distributed delays is obtained. Then, within different normed spaces, several sufficient conditions guaranteeing the existence and uniqueness of an APS are proposed using two fixed-point theorems. Based on the continuity property and some inequality techniques, two theorems insuring the global stability of the unique APS are given. Comparing with known literatures, all conclusions are drawn with slacker restrictions, e.g., do not require the integral of the kernel function determining the distributed delays from zero to positive infinity to be one, and the activation functions to be bounded, etc.; besides, all criteria are obtained by different ways. Finally, two illustrative examples show the validity and that all criteria are easy to check and apply

Letters: On the almost periodic solution of generalized Hopfield neural networks with time-varying delays

This paper presents several sufficient conditions about existence, uniqueness and stability of the almost periodic solution of general Hopfield neural networks with time-varying delays using exponential dichotomy, several fixed point theorems, Halanay inequality, Lyapunov functional and some inequality techniques. These results extend and improve some known relevant works, e.g. the restrictions to the connection weight matrices are slacker, and it is not required that the activation functions are globally Lipschitzian. Most importantly, these conditions are easy to check and apply. Finally, one example is employed to illustrate the conclusions, and the simulated results show the validity. Particularly, the right assertion about the existence, uniqueness and stability of the almost periodic solution of the specific generalized Hopfield neural networks is given only by our criteria, and the relevant criteria provided by a recent reference fail.

Letter: A simple functional neural network for computing the largest and smallest eigenvalues and corresponding eigenvectors of a real symmetric matrix

Efficient computation of the largest eigenvalue and the smallest eigenvalue of a real symmetric matrix is a very important problem in engineering. Using neural networks to complete these operations is in an asynchronous manner and can achieve high performance. This paper proposes a concise functional neural network (FNN) expressed as a differential equation and designs steps to do this work. Firstly, the mathematical analytic solution of the equation is received, and then the convergence properties of this FNN are fully gained. Finally, the computing steps are designed in detail. The proposed method can compute the smallest eigenvalue and the largest eigenvalue whether the matrix is non-definite, positive definite or negative definite. Compared with other methods based on neural networks, this FNN is very simple and concise, so it is very easy to realize.

Letter: A functional neural network for computing the largest modulus eigenvalues and their corresponding eigenvectors of an anti-symmetric matrix

Efficient computation of the largest modulus eigenvalues of a real anti-symmetric matrix is a very important problem in engineering. Using a neural network to complete these operations is in an asynchronous manner and can achieve high performance. This paper proposes a functional neural network (FNN) that can be transformed into a complex differential equation to do this work. Firstly, the mathematical analytic solution of the equation is received, and then the convergence properties of this FNN are analyzed. The simulation result indicates that with general initial complex values, the network will converge to the complex eigenvector corresponding to the eigenvalue whose imaginary part is positive, and modulus is the largest of all eigenvalues. Comparing with other neural networks used for computing eigenvalues and eigenvectors, this network is adaptive to real anti-symmetric matrices for completing these operations.

Robust Prostate Segmentation Using Intrinsic Properties of TRUS Images

Accurate segmentation is usually crucial in transrectal ultrasound (TRUS) image based prostate diagnosis; however, it is always hampered by heavy speckles. Contrary to the traditional view that speckles are adverse to segmentation, we exploit intrinsic properties induced by speckles to facilitate the task, based on the observations that sizes and orientations of speckles provide salient cues to determine the prostate boundary. Since the speckle orientation changes in accordance with a statistical prior rule, rotation-invariant texture feature is extracted along the orientations revealed by the rule. To address the problem of feature changes due to different speckle sizes, TRUS images are split into several arc-like strips. In each strip, every individual feature vector is sparsely represented, and representation residuals are obtained. The residuals, along with the spatial coherence inherited from biological tissues, are combined to segment the prostate preliminarily via graph cuts. After that, the segmentation is fine-tuned by a novel level sets model, which integrates 1) the prostate shape prior, 2) dark-to-light intensity transition near the prostate boundary, and 3) the texture feature just obtained. The proposed method is validated on two 2-D image datasets obtained from two different sonographic imaging systems, with the mean absolute distance on the mid gland images only 1.06±0.53 mm and 1.25±0.77 mm, respectively. The method is also extended to segment apex and base images, producing competitive results over the state of the art.

Complex-Valued Filtering Based on the Minimization of Complex-Error Entropy

In this paper, we consider the training of complex-valued filter based on the information theoretic method. We first generalize the error entropy criterion to complex domain to present the complex error entropy criterion (CEEC). Due to the difficulty in estimating the entropy of complex-valued error directly, the entropy bound minimization (EBM) method is used to compute the upper bounds of the entropy of the complex-valued error, and the tightest bound selected by the EBM algorithm is used as the estimator of the complex-error entropy. Then, based on the minimization of complex-error entropy (MCEE) and the complex gradient descent approach, complex-valued learning algorithms for both the (linear) transverse filter and the (nonlinear) neural network are derived. The algorithms are applied to complex-valued linear filtering and complex-valued nonlinear channel equalization to demonstrate their effectiveness and advantages.

An adaptive bimodal recognition framework using sparse coding for face and ear

In this paper, we propose an adaptive face and ear based bimodal recognition framework using sparse coding, namely ABSRC, which can effectively reduce the adverse effect of degraded modality. A unified and reliable biometric quality measure based on sparse coding is presented for both face and ear, which relies on the collaborative representation by all classes. For adaptive feature fusion, a flexible piecewise function is carefully designed to select feature weights based on their qualities. ABSRC utilizes a two-phase sparse coding strategy. At first, face and ear features are separately coded on their associated dictionaries for individual quality assessments. Secondly, the weighted features are concatenated to form a unique feature vector, which is then coded and classified in multimodal feature space. Experiments demonstrate that ABSRC achieves quite encouraging robustness against image degeneration, and outperforms many up-to-date methods. Very impressively, even when query sample of one modality is extremely degraded by random pixel corruption, illumination variation, etc., ABSRC can still get performance comparable to the unimodal recognition based on the other modality. A refining sparsity-based biometric quality measure suitable for both face and ear.Flexible piecewise feature weight function can better cope with data degeneration.A two-phase sparse coding strategy facilitates precise quality assessments.Considerable illustrations of quality-based fusion and comprehensive experiments.

Fractional Partial Differential Equation: Fractional Total Variation and Fractional Steepest Descent Approach-Based Multiscale Denoising Model for Texture Image

The traditional integer-order partial differential equation-based image denoising approaches often blur the edge and complex texture detail; thus, their denoising effects for texture image are not very good. To solve the problem, a fractional partial differential equation-based denoising model for texture image is proposed, which applies a novel mathematical method—fractional calculus to image processing from the view of system evolution. We know from previous studies that fractional-order calculus has some unique properties comparing to integer-order differential calculus that it can nonlinearly enhance complex texture detail during the digital image processing. The goal of the proposed model is to overcome the problems mentioned above by using the properties of fractional differential calculus. It extended traditional integer-order equation to a fractional order and proposed the fractional Green’s formula and the fractional Euler-Lagrange formula for two-dimensional image processing, and then a fractional partial differential equation based denoising model was proposed. The experimental results prove that the abilities of the proposed denoising model to preserve the high-frequency edge and complex texture information are obviously superior to those of traditional integral based algorithms, especially for texture detail rich images.