Speech deconvolution as an inverse problem

In this paper, the problem of speech deconvolution is solved. This problem is encountered in limited-bandwidth speech communication systems such as telephone systems. Three solutions are presented for this problem. In the first solution, a Linear Minimum Mean Square Error (LMMSE) approach is used. The necessary assumptions required to reduce the computational complexity of the LMMSE solution are presented. In the second solution, an inverse filter deconvolution approach is presented. Finally, the regularization theory is used to solve this problem. The common thread between all these solutions is that they treat the speech deconvolution problem as an inverse problem considering the speech degradation model. Simulation results reveal the superiority of these solutions for solving the speech deconvolution problem.     

基于特征的文档图像高速插值算法

提出了一种基于文档图像特征的混合插值算法。该算法通过分析文档图像的像素邻域特征,使用邻近插值、双线性插值和双三次样条插值方法估计插值像素灰度值。其中,三次样条插值方法使用了优化算法,即三次样条卷积模板的整数化运算,其时间复杂度接近双线性插值。给出了算法的实现过程。实验结果表明,该算法输出的图像质量接近双三次样条插值,运行效率高于双线性插值算法,在文档图像扫描过程取得了良好的应用效果。     

Efficient implementation of chaotic image encryption in transform domains

The primary goal of this paper is security management in data image transmission and storage. Because of the increased use of images in industrial operations, it is necessary to protect the secret data of the image against unauthorized access. In this paper, we introduce a novel approach for image encryption based on employing a cyclic shift and the 2-D chaotic Baker map in different transform domains. The Integer Wavelet Transform (IWT), the Discrete Wavelet Transform (DWT), and the Discrete Cosine Transform (DCT) are exploited in the proposed encryption approach. The characteristics of the transform domains are studied and used to carry out the chaotic encryption. A comparison study between the transform-domain encryption approaches in the presence of attacks shows the superiority of encryption in the DWT domain.     

Efficient implementation of the image rotation method using chirp Z-transform

The implementation of an image rotation method based on the chirp Z-transform is presented. Low computational costs of the algorithm are achieved by performing the chirp Z-transform using convolution in the Fourier domain. The series of numerical experiments supports the preservation of the image quality, even for multiple rotations.     

Automatic construction of active appearance models as an image coding problem

The automatic construction of active appearance models (AAMs) is usually posed as finding the location of the base mesh vertices in the input training images. In this paper, we repose the problem as an energy-minimizing image coding problem and propose an efficient gradient-descent algorithm to solve it.     

On an inverse problem in optimal control

The problem of optimizing linear-stationary control systems subject to quadratic performance indices and bounded control effort constraints is studied. This problem was first proposed by Letov [1]. His solution, however, has been found to be valid only in some special cases [3]. In this paper, necessary and sufficient conditions for the existence of the optimal control laws of the saturation type are developed. It is shown that the weighting factors in the performance index have to be chosen in a particular manner, and the plants have to satisfy certain conditions to yield an optimal saturation type control law. As the weighting on the control effort in the performance index is decreased, the optimal solution approaches a bang-bang type singular control.     

Computation of natural frequencies and mode shapes of arch dams as an inverse problem

In this paper, a procedure is developed which can be used to identify the natural frequencies and natural modes of an Arch-Dam in a vacuum, from forced vibration testing data of partially filled reservoir, throughout removing the effect of hydrodynamic, by using an effective algorithm. The resonance frequency and the resonance mode of the dam–reservoir system have been computed from the in situ tested data. The solution strategy is verified by studying the dynamic response of simple structures, like beam with the fluid interaction effect taken into account, as the analytical results of its modal properties were available for comparison. The hydrodynamic pressure of the reservoir is calculated using the boundary element method. In the inverse problem procedure, the calculated resonance replaces the measured resonance, which has been obtained practically throughout the in situ testing. In this paper the strategies are extended to analyse the dynamic beam–fluid interaction problem with compressible water. The results derived by solving an inverse problem are compared with the exact analytical responses of the beam.     

Video image scaling technology based on adaptive interpolation algorithm and TTS FPGA implementation

With the rapid development of digital multimedia terminals, the scaling of video images has solved the needs of high-end displays for low-resolution signals and high-definition signals displayed by low-end displays. The main purpose of this article is to convert the input video image signal into a video image signal with the required resolution through video image scaling. This article mainly studies various image scaling algorithms, analyzes their advantages and disadvantages, and adopts a top-down design method to implement FPGA video scaling design and verify the correctness of the results. This article introduces various traditional image interpolation algorithms. On the basis of comparing the subjective and objective quality of the image after the summation of various algorithms, the fuzzy interpolation algorithm is used to zoom the video image. In FPGA design, horizontal scaling and vertical scaling are handled separately, which reduces the computational complexity of image upgrades and facilitates control logic and specific implementation. The experimental results of this paper show that the maximum working frequency that the designed image scaling unit can reach is 153.12 MHz, which can process 1080p video signals in real time. Video image scaling technology has broad market prospects.     

Numerical validation for an inverse matrix eigenvalue problem

We describe an algorithm with which one can verify solutions of an additive inverse matrix eigenvalue problem. The algorithm is based on Newton's method using a new criterion for terminating the iteration. In addition, it yields tight interval bounds for the solutions of the problem, thus guaranteeing most of their leading digits in a given floating point system.     

Efficient order processing in an inverse order picking system

An inverse order picking system inverts the basic logic of traditional picker-to-parts systems where pickers successively visit all shelves storing requested stock keeping units (SKUs). Instead, the picker successively moves bins each containing a particular SKU along a line of multiple order bins and puts items into all bins that require the current SKU. In this setting, we aim at a synchronization between the batches of picking orders concurrently assembled and the sequence of SKUs moved along the line, such that the number of line passings to be accomplished by the picker is minimized. We formalize the resulting optimization problem, prove computational complexity, and derive suited solution procedures. In our computational study, we also address important managerial aspects, such as the sizing of the picking area that restricts the number of picking orders concurrently processed.     

Identification of solitary-wave solutions as an inverse problem: Application to shapes with oscillatory tails

The propagation of stationary solitary waves on an infinite elastic rod on elastic foundation equation is considered. The asymptotic boundary conditions admit the trivial solution along with the solution of type of solitary wave, which is a bifurcation problem.     

联合双边图像插值

传统的图像插值算法及其改进算法都存在边缘模糊、边缘锯齿或局部扭曲变形等问题。提出联合双边图像插值,首先采用传统插值算法对原始图像进行插值,将其结果作为导引图像对原始图像进行联合双边上采样,并给出其核函数方差的自适应确定算法。实验结果表明,与传统插值算法相比,该算法明显改进插值图像质量,与已有的绝大多数非线性改进算法相比还提高了插值速度。     

Solution of an inverse dynamics problem using an eigenvalue reduction technique

The inverse dynamics problem is one in which measurements are made on some of the state variables and it is desired to find the unknown forcing function. This paper describes the use of an eigenvalue reduction technique to reduce the order of the system, in conjunction with dynamic programming, to estimate the forcing function. Several numerical experiments were performed to ascertain the effects of noise, the weighting parameters, and the reduction of modes on the solution.     

Efficient FPGA implementation of DWT and modified SPIHT for lossless image compression

In this paper, we present an implementation of the image compression technique set partitioning in hierarchical trees (SPIHT) in programmable hardware. The lifting based Discrete Wavelet Transform (DWT) architecture has been selected for exploiting the correlation among the image pixels. In addition, we provide a study on what storage elements are required for the wavelet coefficients. A modified SPIHT (Set Partitioning in Hierarchical Trees) algorithm is presented for encoding the wavelet coefficients. The modifications include a simplification of coefficient scanning process, use of a 1-D addressing method instead of the original 2-D arrangement for wavelet coefficients and a fixed memory allocation for the data lists instead of the dynamic allocation required in the original SPIHT. The proposed algorithm has been illustrated on both the 2-D Lena image and a 3-D MRI data set and is found to achieve appreciable compression with a high peak-signal-to-noise ratio (PSNR).     

A new hyper-heuristic as a general problem solver: an implementation in HyFlex

This study provides a new hyper-heuristic design using a learning-based heuristic selection mechanism together with an adaptive move acceptance criterion. The selection process was supported by an online heuristic subset selection strategy. In addition, a pairwise heuristic hybridization method was designed. The motivation behind building an intelligent selection hyper-heuristic using these adaptive hyper-heuristic sub-mechanisms is to facilitate generality. Therefore, the designed hyper-heuristic was tested on a number of problem domains defined in a high-level framework, i.e., HyFlex. The framework provides a set of problems with a number of instances as well as a group of low-level heuristics. Thus, it can be considered a good environment to measure the generality level of selection hyper-heuristics. The computational results demonstrated the generic performance of the proposed strategy in comparison with other tested hyper-heuristics composed of the sub-mechanisms from the literature. Moreover, the performance and behavior analysis conducted for the hyper-heuristic clearly showed its adaptive characteristics under different search conditions. The principles comprising the here presented algorithm were at the heart of the algorithm that won the first international cross-domain heuristic search competition.     

图像插值技术综述

图像插值是图像处理中最基本的技术之一,得到了广泛研究和应用.将图像插值技术分为传统插值、基于边缘的插值和基于区域的插值3类,介绍了各类技术的一般实现方法和典型算法;从原理上分析了各类算法的科学性;并对不同算法进行了对比实验和讨论.实验结果表明,基于区域的插值方法原理更科学,插值图像的主观和客观质量最好;最后,给出了图像插值的发展形式及前景.     

Some results on an inverse problem in multivariable systems

A systems property of great interest to the control engineer concerns the system's ability to exhibit those responses which are asked of it. This capability is governed by the nature of the system equations and the set of inputs which are available to 'force' the system. Given the set of desired system responses Sy, and the system equations, this paper is concerned with characterizing the set of inputs which suffice to generate Sy.     

Dynamic programming and an inverse problem in transport theory

Summary The neutron theory of transport processes generally is devoted to problems of determining properties of radiation fields, given certain sources and transporting media. Nevertheless, it has long been recognized that inverse problems, in which we measure some aspects of the radiation field and wish to deduce properties of the sources and the media, are of great significance in the areas of neutron transport, radiative transfer in planetary atmospheres, and astrophysics, among others.In previous papers we have studied the inverse problem using quasilinearization. Here, we wish to indicate the possibility of the application of dynamic programming. We discuss the possibility of reducing the dimensionality of the computational process and provide some numerical results.

---

Zusammenfassung Die Theorie der Transportprozesse von Neutronen befaßt sich gewöhnlich mit Eigenschaften von Strahlungsfeldern, wenn bestimmte Quellen und Transportmedien vorgegeben sind.Es ist jedoch schon seit langer Zeit bekannt, daß inverse Probleme, bei denen bestimmte Größen des Strahlungsfeldes gemessen werden und daraus Eigenschaften der Quellen und Medien abgeleitet werden sollen, von großer Wichtigkeit sind in den Gebieten des Neutronentransports, des Strahlungsübergangs in der Atmosphäre von Planeten, der Astrophysik u. a.In früheren Arbeiten haben wir das inverse Problem mittels Quasilinearisation studiert. Hier wollen wir die Möglichkeit andeuten, die dynamische Programmierung anzuwenden. Wir diskutieren die Möglichkeit, den Rechenaufwand zu reduzieren und geben einige numerische Resultate an.

---

---

With 1 Figure

---

This research is sponsored by the United States Air Force under Project RAND-contract No. AF 49 (638)-700 monitored by the Directorate of Development Plants, Deputy Chief of Staff, Research and Development, Hq USAF. Views or conclusions contained herein should not be interpreted as representing the official opinion or policy of the United States Air Force.     

Nonlinear regulator theory and an inverse optimal control problem

Nonlinear optimal regulators are discussed, and some useful properties are isolated. An inverse problem of nonlinear regulator design is posed and solved.     

逆模型下的高效声源定位算法研究

与波束形成算法相比,广义互相关逆模型宽带声源定位算法提供了空间的高分辨率,但需要更高的计算量。为了提升广义互相关逆模型算法的计算效率,并尽可能保留其分辨率优势,本文提出了一种高效的声源定位算法。该算法首先去除麦克风阵列输出功率较小的网格点来压缩计算网格,其次,在几何因素条件下利用基于密度的聚类进一步压缩网格,最后仅用保留下的点进行计算,从而大大降低了计算规模。真实数据实验证明,本文所提算法能有效提升广义互相关逆模型算法的计算效率。