An efficient method for estimating time-dependent global reliability sensitivity

Structural and Multidisciplinary Optimization - Time-dependent global reliability sensitivity can quantify the effect of input variables in their whole distribution ranges on the time-dependent...     

An efficient deconvolution algorithm for estimating oxygen consumption during muscle activities

The reconstruction of an unknown input function from noisy measurements in a biological system is an ill-posed inverse problem. Any computational algorithm for its solution must use some kind of regularization technique to neutralize the disastrous effects of amplified noise components on the computed solution. In this paper, following a hierarchical Bayesian statistical inversion approach, we seek estimates for the input function and regularization parameter (hyperparameter) that maximize the posterior probability density function. We solve the maximization problem simultaneously for all unknowns, hyperparameter included, by a suitably chosen quasi-Newton method. The optimization approach is compared to the sampling-based Bayesian approach. We demonstrate the efficiency and robustness of the deconvolution algorithm by applying it to reconstructing the time courses of mitochondrial oxygen consumption during muscle state transitions (e.g., from resting state to contraction and recovery), from the simulated noisy output of oxygen concentration dynamics on the muscle surface. The model of oxygen transport and metabolism in skeletal muscle assumes an in vitro cylindrical structure of the muscle in which the oxygen from the surrounding oxygenated solution diffuses into the muscle and is then consumed by the muscle mitochondria. The algorithm can be applied to other deconvolution problems by suitably replacing the forward model of the system.     

An efficient visual tracking method for multiple moving targets

An efficient algorithm of the edge detection according to integrating the edge gradient with the average filter is proposed, which can significantly reduce sensitivity of the background subtraction method to noise and illumination. Taking into account the features of the target such as color, size, etc., a new modified Nearest Neighbor （NN） algorithm for data association using the target features is designed. A designed Interacting Multiple Model （IMM） filter is utilized to track the maneuvering target motion, i.e. the feature point （called the centroid of the target） motion of the target. The algorithms are validated via an example with natural video sequences. The results show the algorithms are performances and validity for visual tracking. In complex environment, the algorithm can still work well.     

An efficient method for estimating null values in relational databases

Generally, a database system containing null value attributes will not operate properly. This study proposes an efficient and systematic approach for estimating null values in a relational database which utilizes clustering algorithms to cluster data, and a regression coefficient to determine the degree of influence between different attributes. Two databases are used to verify the proposed method: (1) Human resource database; and (2) Waugh's database. Furthermore, the mean of absolute error rate (MAER) and average error are used as evaluation criteria to compare the proposed method with other methods. It demonstrates that the proposed method is superior to existing methods for estimating null values in relational database systems. Jia-Wen Wang was born on September 5, 1978, in Taipei, Taiwan, Republic of China. She received the M.S. degree in information management from the National Yunlin University of Science and Technology, Yunlin, Taiwan, in 2003. Since 2003, she has been a PhD degree student in Information Management Department at the National Yunlin University of Science and Technology. Her current research interests include fuzzy systems, database systems, and artificial intelligence. Ching-Hsue Cheng received the B.S. degree in mathematics from Chinese Military Academy, Taiwan, in 1982, the M.S. degree in applied mathematics from the Chung Yuan Christian University, Taiwan, in 1988, and the Ph.D. degree in system engineering and management from National Defence University, Taiwan, in 1994. Currently, he is a professor of the Department of Information Management, National YunLin University of Technology & Science. His research interests are in decision science, soft computing, software reliability, performance evaluation, and fuzzy time series. He has published more than 120 refereed papers in these areas. He has been a principal investigator and project leader in a number of projects with government, and other research-sponsoring agencies.     

An effective pseudospectral method for constraint dynamic optimisation problems with characteristic times

Dynamic optimisation problem with characteristic times, widely existing in many areas, is one of the frontiers and hotspots of dynamic optimisation researches. This paper considers a class of dynamic optimisation problems with constraints that depend on the interior points either fixed or variable, where a novel direct pseudospectral method using Legendre–Gauss (LG) collocation points for solving these problems is presented. The formula for the state at the terminal time of each subdomain is derived, which results in a linear combination of the state at the LG points in the subdomains so as to avoid the complex nonlinear integral. The sensitivities of the state at the collocation points with respect to the variable characteristic times are derived to improve the efficiency of the method. Three well-known characteristic time dynamic optimisation problems are solved and compared in detail among the reported literature methods. The research results show the effectiveness of the proposed method.     

流程雁阵的多目标跟踪优化方法

流程雁阵(Process Goose Queue,PGQ)[1]为流程系统的分解协调优化提供了一个新的方法,然而,目前PGQ方法尚存在许多不足,如简单地将个体PGQ的状态跟踪处理为单目标优化问题,这与实际流程操作不符;而多级PGQ系统优化仍采用传统的数学规划方法,对模型要求苛刻且依赖于初值的选取。为此,论文提出了一个面向流程雁阵多目标跟踪的优化方法。首先对多级PGQ系统进行了结构优化,然后将NSGA-Ⅱ用于多级PGQ系统中个体PGQ的多目标优化求解,得到Pareto解集;在此基础上,将逼近理想解捧序法(TOPSIS)和扩展傅里叶幅值灵敏度分析法(EFAST)应用于个体PGQ的多目标决策,并从Pareto解集选取最优解在多级PGQ系统中逐级传递,实现流程系统的分解协调优化。仿真实例验证了方法的可行性和有效性。     

An identification method for estimating the inertia parameters of a manipulator

This article presents an off-line identification method to estimate the minimal knowledge of the inertia parameters for determining the dynamic model of a manipulator. A new approach is proposed to find a set of the minimal knowledge of the inertia parameters. This set is recursively estimated by moving one joint at a time. The off-line identification procedure also provides a sufficient condition for a persistently exciting trajectory. A simulation example of Stanford arm illustrates the validity and simplicity of the identification procedure.     

A particle swarm optimisation algorithm with interactive swarms for tracking multiple targets

We propose a novel particle swarm optimisation algorithm that uses a set of interactive swarms to track multiple pedestrians in a crowd. The proposed method improves the standard particle swarm optimisation algorithm with a dynamic social interaction model that enhances the interaction among swarms. In addition, we integrate constraints provided by temporal continuity and strength of person detections in the framework. This allows particle swarm optimisation to be able to track multiple moving targets in a complex scene. Experimental results demonstrate that the proposed method robustly tracks multiple targets despite the complex interactions among targets that lead to several occlusions.     

An interacting Fuzzy-Fading-Memory-based Augmented Kalman Filtering method for maneuvering target tracking

In this paper, the interaction and combination of Fuzzy Fading Memory (FFM) technique and Augmented Kalman Filtering (AUKF) method are presented for the state estimation of non-linear dynamic systems in presence of maneuver. It is shown that the AUKF method in conjunction with the FFM technique (FFM-AUKF) can estimate the target states appropriately since the FFM tunes the covariance matrix of the AUKF method in presence of unknown target accelerations by using a fuzzy system. In addition, the benefits of both FFM technique and AUKF method are employed in the scheme of well-known Interacting Multiple Model (IMM) algorithm. The proposed Fuzzy IMM (FIMM) algorithm does not need the predefinition and adjustment of sub-filters with respect to the target maneuver and reduces the number of required sub-filters to cover the wide range of unknown target accelerations. The Monte Carlo simulation analysis shows the effectiveness of the above-mentioned methods in maneuvering target tracking.     

An implicit relevance feedback method for CBIR with real-time eye tracking

Relevance feedback is an efficient approach to improve the performance of content-based image retrieval systems, and implicit relevance feedback approaches, which gather users’ feedback by biometric devices (e.g. eye tracker), have extensively investigated in recent years. This paper proposes a novel image retrieval system with implicit relevance feedback, named eye tracking based relevance feedback system (ETRFs). ETRFs is composed of three main modules: image retrieval subsystem based on bag-of-word architecture; user relevance assessment that implicitly acquires relevant images with the help of a modern eye tracker; and relevance feedback module that applies a weighted query expansion method to fuse users’ relevance feedback. ETRFs is implemented online and real-time, which makes it remarkably distinguish from other offline systems. Ten subjects participate our experiments on the dataset of Oxford buildings and UKBench. The experimental results demonstrate that ETRFs achieves notable improvement for image retrieval performance.     

Markerless human articulated tracking using hierarchical particle swarm optimisation

In this paper, we address markerless full-body articulated human motion tracking from multi-view video sequences acquired in a studio environment. The tracking is formulated as a multi-dimensional non-linear optimisation and solved using particle swarm optimisation (PSO), a swarm-intelligence algorithm which has gained popularity in recent years due to its ability to solve difficult non-linear optimisation problems. We show that a small number of particles achieves accuracy levels comparable with several recent algorithms. PSO initialises automatically, does not need a sequence-specific motion model and recovers from temporary tracking divergence through the use of a powerful hierarchical search algorithm (HPSO). We compare experimentally HPSO with particle filter (PF), annealed particle filter (APF) and partitioned sampling annealed particle filter (PSAPF) using the computational framework provided by Balan et al. HPSO accuracy and consistency are better than PF and compare favourably with those of APF and PSAPF, outperforming it in sequences with sudden and fast motion. We also report an extensive experimental study of HPSO over ranges of values of its parameters.     

Post-stratified calibration method for estimating quantiles

The estimation of quantiles in the presence of auxiliary information is discussed. Calibration and poststratification techniques provide simple and practical procedures for incorporating auxiliary information into the estimation of distribution functions, which can offer some useful gains in efficiency. The estimator proposed combines these techniques and possesses a number of desirable properties, including yielding a genuine distribution function, providing simplicity of computation and generalizing Silva and Skinner’s estimator. This proposed procedure is compared to alternative methods. On the basis of simulation studies, the proposed post-stratified calibration estimator presents a good level of performance and comprises a valid alternative to other estimators of the distribution function.     

An optimisation strategy for industrial metal forming processes

Product improvement and cost reduction have always been important goals in the metal forming industry. The rise of finite element (FEM) simulations for processes has contributed to these goals in a major way. More recently, coupling FEM simulations to mathematical optimisation techniques has shown the potential to make a further giant contribution to product improvement and cost reduction. Much research on the optimisation of metal forming processes has been published during the last couple of years. Although the results are impressive, the optimisation techniques are generally only applicable to specific optimisation problems for specific products and specific metal forming processes. As a consequence, applying optimisation techniques to other metal forming problems requires a lot of optimisation expertise, which forms a barrier for more general industrial application of these techniques. In this paper, we overcome this barrier by proposing a generally applicable optimisation strategy that makes use of FEM simulations of metal forming processes. It consists of a structured methodology for modelling optimisation problems related to metal forming. Subsequently, screening is applied to reduce the size of the optimisation problem by selecting only the most important design variables. Finally, the reduced optimisation problem is solved by an efficient optimisation algorithm. The strategy is generally applicable in a sense that it is not constrained to a certain type of metal forming problem, product or process. Also, any FEM code may be included in the strategy. Furthermore, the structured approach for modelling and solving optimisation problems should enable non-optimisation specialists to apply optimisation techniques to improve their products and processes. The optimisation strategy has been successfully applied to a hydroforming process, which demonstrates the potential of the optimisation of metal forming processes in general and more specific the proposed optimisation strategy.     

Linearisation: An optimisation for nonlinear functional programs

Functional programming languages have great appeal from the point of view of both software design and amenability to formal reasoning, but to date they have suffered from poor performance when run on conventional computers. A promising solution to this problem may be provided by program transformation and several schemes have been proposed which can give quite impressive optimisations. However, these are at best only semi-automatic, requiring reasoning on behalf of the programmer to assist the transformation process. Part of the problem is that these schemes must take into account not only functions but also the objects to which they are applied in the defining expressions. By reasoning at the function level, the auxiliary domain of objects need not be considered explicitly, and transformations can be derived in terms of identities between functional expressions, rather than via sets of equations satisfied by objects from a certain class.

By expressing functional expressions in variable-free form, we use algebraic methods, based on the functional algebra of the language FP, to transform a certain class of nonlinear functions into linear form. A function in this class generates a reduction graph in the form of a balanced tree when applied to an argument, whereas a linear function generates a single-spine tree and so executes with a number of function calls which is linear in the size of its argument. Thus, for example, tail recursive functions form a small subset of the class of linear functions. Further optimisations include the tupling of functions which are defined by mutal recursion, and we identify conditions under which these are equivalent to a linear function. The compiler is able to detect if the conditions required by these transformation theorems are satisfied, and generate the appropriate optimised functions.     

An improved NDVI-based threshold method for estimating land surface emissivity using MODIS satellite data

This work estimated the land surface emissivities (LSEs) for MODIS thermal infrared channels 29 (8.4–8.7 μm), 31 (10.78–11.28 μm), and 32 (11.77–12.27 μm) using an improved normalized difference vegetation index (NDVI)-based threshold method. The channel LSEs are expressed as functions of atmospherically corrected reflectance from the MODIS visible and near-infrared channels with wavelengths ranging from 0.4 to 2.2 μm for bare soil. To retain the angular information, the vegetation LSEs were explicitly expressed in the NDVI function. The results exhibited a root mean square error (RMSE) among the estimated LSEs using the improved method, and those calculated using spectral data from Johns Hopkins University (JHU) are below 0.01 for channels 31 and 32. The MODIS land surface temperature/emissivity (LST/E) products, MOD11_L2 with LSE derived via the classification-based method with 1 km resolution and MOD11C1 with LSE retrieved via the day/night LST retrieval method at 0.05° resolution, were used to validate the proposed method. The resultant variances and entropies for the LSEs estimated using the proposed method were larger than those extracted from MOD11_L2, which indicates that the proposed method better described the spectral variation for different land covers. In addition, comparing the estimated LSEs to those from MOD11C1 yielded RMSEs of approximately 0.02 for the three channels; however, more than 70% of pixels exhibited LSE differences within 0.01 for channels 31 and 32, which indicates that the proposed method feasibly depicts LSE variation for different land covers.     

An algorithm for tracking multiple targets

An algorithm for tracking multiple targets in a cluttered enviroment is developed. The algorithm is capable of initiating tracks, accounting for false or missing reports, and processing sets of dependent reports. As each measurement is received, probabilities are calculated for the hypotheses that the measurement came from previously known targets in a target file, or from a new target, or that the measurement is false. Target states are estimated from each such data-association hypothesis using a Kalman filter. As more measurements are received, the probabilities of joint hypotheses are calculated recursively using all available information such as density of unknown targets, density of false targets, probability of detection, and location uncertainty. This branching technique allows correlation of a measurement with its source based on subsequent, as well as previous, data. To keep the number of hypotheses reasonable, unlikely hypotheses are eliminated and hypotheses with similar target estimates are combined. To minimize computational requirements, the entire set of targets and measurements is divided into clusters that are solved independently. In an illustrative example of aircraft tracking, the algorithm successfully tracks targets over a wide range of conditions.     

An efficient method for tracking failure detection using parallel correlation filtering and Siamese network

Applied Intelligence - In recent years, object tracking based on Siamese network has attracted the attention by taking advantage of both speed and accuracy. However, when similar object...