基于小波滤波器组的光流估计

为了解决光流估计时存在的孔径问题和时域混叠，在Bernard的理论基础上，给出了基于快速滤波器组的方法：小波函数在频域可以表示为滤波器组的乘积．所以时域的小波系数由傅里叶变换转换到频域。通过滤波器组快速求解。由于解析小波具有实小波所没有的优点．采用解析小波计算光流。文中给出该方法的结果。还给出了Horn-Schunck的正则化方法的结果。比较后可知，本文的方法能较好地对运动物体进行光流估计。     

基于QPSO的两通道正交镜像滤波器组的优化设计*

通过分析两通道正交镜像滤波器组的重构条件,提出了一种新的两通道正交镜像滤波器(QMF)组的设计方法,将完全重构的两通道QMF组的设计问题转换为由原型低通滤波器在通带、阻带和过渡带幅频响应的均方误差加权和组成的一种无约束的非线性优化问题,用量子粒子群优化算法(QPSO)来有效求解该类优化问题,获得该问题的全局最优解.最后...     

Design of shift-invariant nonuniform cosine-modulated filter bank with arbitrary integer sampling factors

Shift-invariance and unequal frequency partition are two desirable properties in many signal/image processing applications. In this paper, we treat a simple but efficient approach to design shift-invariant nonuniform cosine-modulated filter bank (SI-NCMFB) with arbitrary integer sampling factors. It consists of two subsystems where the first one is obtained by merging the consecutive subbands of a cosine-modulated filter bank, and the second one is derived under the condition of shift-invariance. By imposing the significant aliasing in the two subsystems to cancel each other structurally, the proposed SI-NCMFB achieves the properties of shift-invariance and flexible frequency partition. Since both the first and second subsystems come from just one prototype filter, the SI-NCMFB design is reduced to a filter design, leading to low design complexity. Furthermore, we extend the SI-NCMFB to two dimensions by separable operations. Unlike traditional separable transforms, the resulting 2D subbands have flexible directional-selectivity which is highly expected in image representation. Several simulation experiments are given to verify the proposed SI-NCMFB and the designed SI-NCMFB is of near-perfect reconstruction.     

Integrated fault detection and control for two‐dimensional Markovian jump systems

This paper is concerned with the problem of integrated fault detection and control for a class of two‐dimensional (2D) discrete‐time Markovian jump systems. The mathematical model of 2D Markovian jump systems is established upon the well‐known Roesser model, and a faults detection filter/controller is proposed to detect faults and meet some control specifications simultaneously. In this strategy, it takes into account both the fault detection objective and the control objective simultaneously through certain performance levels. The integrated design problem is then formulated as a multi‐objective optimization problem, which is nonconvex in essence. Furthermore, a two‐step algorithm is developed to solve this nonconvex problem. Sufficient conditions for existence of the desired fault detection filter/controller are established in terms of LMIs. A numerical example is used to demonstrate the effectiveness of the proposed method.     

滤波器组框架理论及其在图信号处理中的应用

滤波器组框架理论是应用数学、信号处理、图像处理和数字通信等领域的重要问题之一,对滤波器组框架的分析和设计问题进行研究有着重要的科学意义和应用前景.近年来,随着高维非规则化数据信息大量涌现,很多学者开始研究图信号处理的滤波器组方法.因此对滤波器组框架理论及其在图信号处理中的应用进行了综述研究.首先对传统滤波器组框架理论的基础知识进行概述,总结滤波器组框架分析与设计方法;然后重点介绍两类图信号处理架构以及图滤波器组的最新研究成果;最后对未来的研究进行展望.     

Marquardt optimization method to design two-channel quadrature mirror filter banks

This paper proposes an efficient method for the design of two-channel, quadrature mirror filter (QMF) bank. The design problem is formulated to minimize an objective function, which is a linear combination of pass-band error and stop-band residual energy of the low-pass analysis filter of the filter bank, and the square error of the overall transfer function of the QMF bank at the quadrature frequency. A new algorithm has been developed to minimize the objective function by optimizing the filter tap weights by using Marquardt optimization method. Simulation results of the algorithm show that the performance of the proposed method is better than that of the already existing methods.     

基于H∞范数优化的干扰观测器的鲁棒设计

研究干扰观测器的鲁棒优化设计方法, 应用H∞范数定义干扰观测器的优化性能评价函数, 把低通滤波器的设计问题转换为H∞闭环回路成形问题. 通过适当处理相对阶次条件等约束, 把带有约束的回路成形问题转换成无约束的H∞标准问题, 然后利用H∞标准问题的求解算法设计滤波器. 此外, 探讨了性能与频率加权函数的关系, 并在此基础上提出了加权函数的选取方法. 实验结果证明该方法设计简单, 且具有最优性和系统性.     

近似完全重构DFT调制滤波器组的设计

提出了一种新的近似完全重构（NPR）离散傅立叶变换（DFT）调制滤波器组的原型滤波器设计方法。其优化的目标函数由滤波器组的输出混叠、输出失真及原型滤波器的阻带能量所导出。在此设计方法中,利用DFT调制滤波器组的特殊结构合理简化了滤波器组的输出混叠,输出失真。原型滤波器的设计最终被转化为一个简单的非线性优化问题。与目前的DFT调制滤波器组的设计方法相比,该方法提供了相当大的灵活性同时具有简单的优化描述。仿真结果证实了所提出的设计方法的有效性。     

Robust fault detection for discrete-time Markovian jump systems with mode-dependent time-delays

This paper investigates a fault detection problem for a class of discrete-time Markovian jump systems with norm-bounded uncertainties and mode-dependent time-delays. Attention is focused on constructing the residual generator based on the filter which its parameters matrices are dependent on the system mode, that is, the fault detection filter is a Markovian jump system as well. The design of fault detection filter is reduced to H-infinity filtering problem by using H-infinity control theory, which can guarantee the difference between the residual and the fault (or, more generally weighted fault) as small as possible in the context of enhancing the robustness of residual to modeling errors, control inputs and unknown inputs. Sufficient condition for the existence of the above filters is established by means of linear matrix inequalities, which can be readily {solved by using standard numerical software. A numerical example is given to illustrate the feasibility of the proposed method.     

Robust fault detection for discrete-time Markovian jump systems with mode-dependent time-delays

This paper investigates a fault detection problem for a class of discrete-time Markovian jump systems with norm-bounded uncertainties and mode-dependent time-delays. Attention is focused on constructing the residual generator based on the filter of which its parameters matrices are dependent on the system mode, that is, the fault detection filter is a Markovian jump system as well. The design of fault detection filter is reduced to H-infinity filtering problem by using H-infinity control theory, which can guarantee the difference between the residual and the fault (or, more generally weighted fault) as small as possible in the context of enhancing the robustness of residual to modeling errors, control inputs and unknown inputs. Sufficient condition for the existence of the above filters is established by means of linear matrix inequalities, which can be readily solved by using standard numerical software. A numerical example is given to illustrate the feasibility of the proposed method.     

Prototype filter design for QAM-based filter bank multicarrier system

In this paper, we provide conditions for a prototype filter design of filter bank multicarrier (FBMC) based on quadrature amplitude modulation (QAM). The conditions consist of a generalized Nyquist criterion (GNC) for nearly perfect reconstruction (NPR) and meeting the stopband condition for a small side-lobe. In a practical environment, a small side-lobe is a key aspect of achieving high spectral efficiency, which becomes an important factor in reducing the size of the guard band among channels. In addition, because the conventional GNC is derived under the assumption of an ideal channel, which can be easily broken over a practical multipath channel, we propose a relaxed NPR that considers the multipath delay using a 2L-oversampled discrete Fourier transform (DFT) in the frequency domain. Based on the relaxation of the GNC depending on a multipath channel, we formulate an optimization problem for a QAM-FBMC prototype filter design and propose a prototype filter with a small side-lobe and reliable BER performance. Simulation results show that the proposed prototype filter remarkably reduces the side-lobe compared to conventional QAM-FBMC based on two types of prototype filters and orthogonal frequency division multiplexing (OFDM) via the trade-off between GNC and side-lobe performance. The benefit of the side-lobe condition allows the proposed prototype filter, the Relaxed-NPR-F, to improve spectral efficiency by reducing the guard band in the frequency domain.     

一类关联时滞组合系统的H_∞滤波

针对当输入噪声为有限能量信号时的渐近稳定的线性滤波器的设计问题,基于Lyapunov稳定性理论,结合线性矩阵不等式技术,提出了一类关联时滞组合系统的H_∞滤波器的设计方案,并利用有界实引理给出了滤波器存在的一个充分条件。为了使得滤波器具有良好的稳态性能,考虑了LMI的优化问题。并将滤波器的设计问题转化为具有线性矩阵不等式约束的凸优化求解问题。通过求解一组LMI,可以得到最优滤波器。     

Multi-dimensional visual tracking using scatter search particle filter

Multi-dimensional visual tracking (MVT) problems include visual tracking tasks where the system state is defined by a high number of variables corresponding to multiple model components and/or multiple targets. A MVT problem can be modeled as a dynamic optimization problem. In this context, we propose an algorithm which hybridizes particle filters (PF) and the scatter search (SS) metaheuristic, called scatter search particle filter (SSPF), where the optimization strategies from SS are embedded into the PF framework. Scatter search is a population-based metaheuristic successfully applied to several complex combinatorial optimization problems. The most representative optimization strategies from SS are both solution combination and solution improvement. Combination stage enables the solutions to share information about the problem to produce better solutions. Improvement stage makes also possible to obtain better solutions by exploring the neighborhood of a given solution. In this paper, we have described and evaluated the performance of the scatter search particle filter (SSPF) in MVT problems. Specifically, we have compared the performance of several state-of-the-art PF-based algorithms with SSPF algorithm in different instances of 2D articulated object tracking problem and 2D multiple object tracking. Some of these instances are from the CVBase’06 standard database. Experimental results show an important performance gain and better tracking accuracy in favour of our approach.     

Metaheuristic algorithms for the design of multiplier-less non-uniform filter banks based on frequency response masking

In this paper, multiplier-less nearly perfect reconstruction tree structured non-uniform filter banks (NUFB) are proposed. When sharp transition width filter banks are to be implemented, the order of the filters and hence the complexity will become very high. The filter banks employ an iterative algorithm which adjusts the cut off frequencies of the prototype filter, to reduce the amplitude distortion. It is found that the proposed design method, in which the prototype filter is designed by the frequency response masking method, gives better results when compared to the earlier reported results, in terms of the number of multipliers when sharp transition width filter banks are needed. To reduce the complexity and power consumption for hardware realization, a design method which makes the NUFB totally multiplier-less is also proposed in this paper. The NUFB is made multiplier-less by converting the continuous filter bank coefficients to finite precision coefficients in the signed power of two space. The filter bank with finite precision coefficients may lead to performance degradation. This calls for the use of suitable optimization techniques. The classical gradient based optimization techniques cannot be deployed here, because the search space consists of only integers. In this context, meta-heuristic algorithm is a good choice as it can be tailor made to suit the problem under consideration. Thus, this design method results in near perfect NUFBs which are simple and multiplier-less and have linear phase and sharp transition width with very low aliasing. Also, different non-uniform bands can be obtained from the tree structured filter bank by rearranging the branches.     

A new approach to design triplet halfband filter banks based on balanced-uncertainty optimization

This paper presents a novel approach to design a class of triplet halfband filter banks (THFB) based on optimized time–frequency localization. First, an Euler–Frobenius polynomial (EFP) is introduced to design a class of halfband polynomial. The vanishing moments and perfect reconstruction conditions are imposed on EFP to obtain maximally flat halfband filter. The resultant halfband filter is optimized using a balanced-uncertainty (BU) metric in order to have balance between time and frequency spread. Next, this optimized halfband filter is used in three-step lifting scheme to obtain analysis and synthesis wavelet filters which have balance between time and frequency localization. The proposed method provides three degrees of freedom that results in flexible design of filter bank. It is observed that the proposed filter bank gives more regularity and better frequency selectivity as compared to existing filter banks. These designed filters are then used for human chromosome image compression application. The performance of the designed filter bank is compared in terms of PSNR using different bit-rates with well-known existing filter banks in order to validate the experimental results.     

A hybrid method for designing linear-phase quadrature mirror filter bank

In this work, a hybrid method in frequency domain for design of a linear-phase quadrature mirror filter bank is proposed. The nonlinear least square hybrid method based on Levenberg–Marquardt (LM) and Quasi-Newton (QN) is developed for the design of a low-pass prototype filter whose responses in the passband and stopband are ideal, and the filter bank response at quadrature frequency is 0.707. The proposed method is employed to optimize the quadratic measure of ideal characteristics of the prototype filter and filter bank at quadrature frequency. Performance and effectiveness of the proposed method in terms of peak reconstruction error (PRE), mean squares error in passband and stopband regions, and error in transition band at quadrature frequency are shown through numerical examples and comparison of the results with other existing method. It was found that the proposed method is very simple and easy to implement for QMF bank design problem.     

Optimized Gabor features for mass classification in mammography

Gabor filter bank has been successfully used for false positive reduction problem and the discrimination of benign and malignant masses in breast cancer detection. However, a generic Gabor filter bank is not adapted to multi-orientation and multi-scale texture micro-patterns present in the regions of interest (ROIs) of mammograms. There are two main optimization concerns: how many filters should be in a Gabor filter band and what should be their parameters. Addressing these issues, this work focuses on finding optimizing Gabor filter banks based on an incremental clustering algorithm and Particle Swarm Optimization (PSO). We employ an SVM with Gaussian kernel as a fitness function for PSO. The effect of optimized Gabor filter bank was evaluated on 1024 ROIs extracted from a Digital Database for Screening Mammography (DDSM) using four performance measures (i.e., accuracy, area under ROC curve, sensitivity and specificity) for the above mentioned mass classification problems. The results show that the proposed method enhances the performance and reduces the computational cost. Moreover, the Wilcoxon signed rank test over the significance level of 0.05 reveals that the performance difference between the optimized Gabor filter bank and non-optimized Gabor filter bank is statistically significant.     

Current state and future directions of multirate filter banks and their applications

This paper provides an overview of the basic concepts, current state and future directions of uniform and nonuniform multirate filter banks and their applications. Research in the areas of multidimensional filter banks, and wavelet transforms, as a direct extension to the topic is still proceeding at a rapid rate. Different design techniques and algorithms that are currently of interest are investigated from hardware complexity and reconstructed signals' quality points of view. Important similarities and differences between different filters categories are noted, which lead to a discussion of important issues that need to be resolved. Distortions and quantization effects present in the filter banks are reviewed and some solutions for controlling different types of distortions are introduced. The problem of designing QMF banks that can provide good frequency resolution while allowing for exact or near perfect reconstruction of the signal is quite challenging, because so many dissimilar types of distortion must be minimized and/or eliminated in the same design context. Recently, the perfect reconstruction cosine-modulated filter bank and filter banks with block sampling have emerged as an attractive choice of filter bank with respect to implementation complexities and design cost. New trends and future directions including selective multirate filter banks, filter banks with nonuniform decimation ratio, filter banks with discrete coefficients, multidimensional filter banks, nonlinear filter banks and data dependent decompositions are presented. Finally, some selected applications of multirate filter banks are described; namely transmultiplexers, speech and image coding, analog voice privacy systems and wavelets based signal and image compression.     

一类网络控制系统的非脆弱耗散滤波器设计

针对网络化控制系统的滤波器参数存在摄动的问题,考虑到传感器-滤波器存在随机时延和丢包,设计对系统待估信号进行估计的加性非脆弱耗散滤波器,其中滤波器参数具有范数有界不确定性。首先使用Lyapunov稳定性理论和线性矩阵不等式分析方法,推导出耗散滤波器存在的充分条件,在此基础上考虑到滤波器参数不确定性,从而将非脆弱耗散滤波问题转化为求解线性矩阵不等式的问题,然后通过求解线性矩阵不等式得到非脆弱耗散滤波器的参数表达式,最后数值仿真验证了设计方法的有效性。