Approximate all nearest neighbor search for high dimensional entropy estimation for image registration

Information theoretic criteria such as mutual information are often used as similarity measures for inter-modality image registration. For better performance, it is useful to consider vector-valued pixel features. However, this leads to the task of estimating entropy in medium to high dimensional spaces, for which standard histogram entropy estimator is not usable. We have therefore previously proposed to use a nearest neighbor-based Kozachenko-Leonenko (KL) entropy estimator.Here we address the issue of determining a suitable all nearest neighbor (NN) search algorithm for this relatively specific task.We evaluate several well-known state-of-the-art standard algorithms based on k-d trees (FLANN), balanced box decomposition (BBD) trees (ANN), and locality sensitive hashing (LSH), using publicly available implementations. In addition, we present our own method, which is based on k-d trees with several enhancements and is tailored for this particular application.We conclude that all tree-based methods perform acceptably well, with our method being the fastest and most suitable for the all-NN search task needed by the KL estimator on image data, while the ANN and especially FLANN methods being most often the fastest on other types of data. On the other hand, LSH is found the least suitable, with the brute force search being the slowest.     

基于加权自学习散列的高维数据最近邻查询算法

因为查询和存储具有高效性,学习型散列逐渐被应用于解决最近邻查询问题.学习型散列将高维数据转化成二进制编码,并使得原始高维空间中越相似的数据对应二进制编码的汉明距离越小.在实际应用中,每次查询都会返回许多与查询点汉明距离相同而编码互不相同的数据.如何对这些数据进行排序是一个难题.提出了一种基于加权自学习散列的近邻查找算法.实验结果表明,算法能够高效地对具有相同汉明距离的不同编码进行重排序,加权排序后查询的F1值约是原来的2倍并优于同系算法,时间开销可比直接计算原始距离进行排序降低一个数量级.     

Penalty function approach in heuristic algorithms for constrained redundancy reliability optimization

To solve the problem of constrained redundancy reliability optimization, several heuristic algorithms have been developed in the literature. Most of these algorithms search for the solutions remaining within the feasible boundary e.g. [15], [20]. Perhaps the most interesting & efficient heuristic algorithm in terms of solution quality is that given by KYA, in which the search is made not only in the feasible region but also into the bounded infeasible region by making an excursion, which returns to the feasible region with a possibly improved solution. In this paper, a heuristic algorithm based on the penalty function approach is proposed to solve the constrained redundancy optimization problem for complex systems. An excursion is made into the infeasible region, but an adaptive penalty function helps the search not to go too far into the infeasible region. Thus, promising feasible & infeasible regions of the search space are explored efficiently & effectively to identify finally an optimal or near optimal solution. Computational experiments are conducted on 11 sets of problems (10 with linear constraints, and 1 with nonlinear constraints); each with 10 different randomly generated initial solutions. Comparison is made between the proposed algorithm P-Alg, N-N algorithm [15], Shi algorithm [20], and KYA [9] . It is observed that P-Alg performs consistently better than others, showing an overall improvement in various measures of performance. Besides, as P-Alg does not require any assumptions on the nature of the objective & constraint functions, it can solve a wide variety of problems.     

A tabu search algorithm for routing optimization in mobile ad-hoc networks

In this paper, we propose a routing optimization algorithm to efficiently determine an optimal path from a source to a destination in mobile ad-hoc networks. To determine an optimal path for the nodes is important for transmitting data between nodes in densely deployed networks. In order to efficiently transmit data to its destination, the appropriate routing algorithms must be implemented in mobile ad-hoc networks. The proposed algorithm is designed by using a tabu search mechanism that is a representative meta-heuristic algorithm. The proposed tabu search algorithm carries out two neighborhood generating operations in order to determine an optimal path and minimize algorithm execution time. We compare the proposed tabu search algorithm with other meta-heuristic algorithms, which are the genetic algorithm and the simulated annealing, in terms of the routing cost and algorithm execution time. The comparison results show that the proposed tabu search algorithm outperforms the other algorithms and that it is suitable for adapting the routing optimization problem.     

A modular approach for simulation-based optimization of MEMS

The importance of MEMS optimization concerning performance, power consumption, and reliability is increasing. A variety of specialized tools is available in the MEMS design flow. FEM tools (ANSYS, CFD-ACE+) are widely used on component level. System level simulations are carried out using simplified models and simulators like Saber or ELDO. There is a lack of simulator-independent optimization support. Only a few simulators offer optimization capabilities.Our approach aims at a flexible combination of simulators and optimization algorithms by partitioning the optimization cycle into simulation, error calculation, optimization and model generation. This new method is translated into a modular optimization system named Moscito implemented in Java. Several optimization algorithms are available: methods without derivatives, methods using derivatives and stochastic approaches. Interfaces to simulators like ANSYS, Saber, MATLAB are implemented. Thus the optimization problem can be handled on different levels of model abstraction (FEM, ordinary differential equations, generalized networks, block diagrams, etc.). A graphical user interface supports control and visualization of the optimization. The modules of the optimization system are able to communicate via the Internet to allow a distributed, web-based optimization.     

A new heuristic approach for redundancy optimization problem

The present paper describes a simple heuristic approach to solve redundancy optimization problem without much of computation to help reliability designers choose an appropriate configuration. The solution technique is simpler than all previous methods and offers optimal results with a high success rate.     

A subspace ensemble framework for classification with high dimensional missing data

Real world classification tasks may involve high dimensional missing data. The traditional approach to handling the missing data is to impute the data first, and then apply the traditional classification algorithms on the imputed data. This method first assumes that there exist a distribution or feature relations among the data, and then estimates missing items with existing observed values. A reasonable assumption is a necessary guarantee for accurate imputation. The distribution or feature relations of data, however, is often complex or even impossible to be captured in high dimensional data sets, leading to inaccurate imputation. In this paper, we propose a complete-case projection subspace ensemble framework, where two alternative partition strategies, namely bootstrap subspace partition and missing pattern-sensitive subspace partition, are developed for incomplete datasets with even missing patterns and uneven missing patterns, respectively. Multiple component classifiers are then separately trained in these subspaces. After that, a final ensemble classifier is constructed by a weighted majority vote of component classifiers. In the experiments, we demonstrate the effectiveness of the proposed framework over eight high dimensional UCI datasets. Meanwhile, we apply the two proposed partition strategies over data sets with different missing patterns. As indicated, the proposed algorithm significantly outperforms existing imputation methods in most cases.     

A new predictive search area approach for fast block motion estimation

According to the observation on the distribution of motion differentials among the motion vector of any block and those of its four neighboring blocks from six real video sequences, this paper presents a new predictive search area approach for fast block motion estimation. Employing our proposed simple predictive search area approach into the full search (FS) algorithm, our improved FS algorithm leads to 93.83% average execution-time improvement ratio, but only has a small estimation accuracy degradation. We also investigate the advantages of computation and estimation accuracy of our improved FS algorithm when compared to the edge-based search algorithm of Chan and Siu (see IEEE Trans. Image Processing, vol.10, p.1223-1238, Aug. 2001); experimental results reveal that our improved FS algorithm has 74.33% average execution-time improvement ratio and has a higher estimation accuracy. Finally, we further compare the performance among our improved FS algorithm, the three-step search algorithm, and the block-based gradient descent search algorithm.     

A general-purpose optimization approach for designing two-channel FIR filterbanks

An efficient general-purpose optimization approach is proposed for designing two-channel finite impulse response (FIR) filterbanks. This technique can be used for optimizing two-channel FIR filterbanks in all alias-free cases proposed in the literature. The generalized problem is to minimize the maximum of the stopband energies of the two analysis filters subject to the given passband and transition band constraints and the given allowable reconstruction error. Therefore, in addition to the perfect-reconstruction filterbanks, nearly perfect-reconstruction banks can be optimized in a controlled manner. The optimization is carried out in two steps. In the first step, for the selected type of the filterbank, a good starting-point filterbank for further optimization is generated using an existing design scheme. The second step involves optimizing the filterbank with the aid of a modified Dutta-Vidyasagar (1977) algorithm. Several examples are included, illustrating the efficiency and the flexibility of the proposed approach.     

Trust-aware energy-efficient stable clustering approach using fuzzy type-2 Cuckoo search optimization algorithm for wireless sensor networks

Wireless Networks - With the advancement of communication and sensor technologies, it has become possible to develop low-cost circuitry to sense and transmit the state of surroundings. Wireless...     

高压输电线异物激光清除三维仿真研究

激光以其优异的性能在输电线路异物清除领域具有广阔的应用前景。为了安全、高效地清除输电线异物,须对激光与异物组织的作用机理进行研究。本文分析讨论了激光与尼龙材料的热作用机理,建立了激光束切割尼龙材料的三维数值仿真模型,仿真研究了不同激光移动速度和光斑尺寸下尼龙材料三维动态烧蚀形貌,分析研究了激光移动速度和光斑尺寸对材料烧蚀深度的影响趋势。结果表明,激光功率和光斑尺寸一定时,材料烧蚀的最高温度和烧蚀深度与激光移动速度成反比;激光功率和移动速度一定时,材料烧蚀的最高温度和烧蚀深度与光斑尺寸成反比。当激光功率为100W、光斑半径1cm时,烧蚀过程中不发生明火的激光最小移动速度为05cm/s。当激光移动速度为2cm/s、功率为100W 时,不发生明火的光斑最小直径为03cm。     

A novel quadrilinear CCD for high resolution line arrays

This paper presents a novel quadrilinear CCD, allowing both serial and parallel clocking with one set of busbars common to inner and outer registers on the same side of the array. This is made possible by the special organization of the charge transfer from inner to outer registers. Line arrays having 384, 808, 1728, and 3456 sensors have been fabricated successfully on a pitch of 8 µm; the sensor nonuniformity is less than ±5 percent and the transfer loss is 1.10^-5in both inner and outer registers.     

A robust approach for high resolution frequency estimation

A robust estimation method for frequencies of received signals is considered. The influence function of a robust estimate is derived for the sinusoidal signals. The variance of the robust estimate is also derived using the influence function. It is found that the robust estimate attains the Cramer-Rao lower bound (CRLB) for the contaminated Gaussian distribution. It is of order O(N^-3) and is close to the CRLB for the perfect Gaussian distribution. The authors introduce some basic definitions for the high-resolution frequency estimation method and prove that the robust estimate has the high-resolution property. This property is also confirmed by numerical simulations     

The genetic search approach. A new learning algorithm for adaptiveIIR filtering

An “evolutionary” approach called the genetic algorithm (GA) was introduced for multimodal optimization in adaptive IIR filtering. However, the disadvantages of using such an algorithm are slow convergence and high computational complexity. Initiated by the merits and shortcomings of the gradient-based algorithms and the evolutionary algorithms, we developed a new hybrid search methodology in which the genetic-type search is embedded into gradient-descent algorithms (such as the LMS algorithm). The new algorithm has the characteristics of faster convergence, global search capability, less sensitivity to the choice of parameters, and simple implementation. The basic idea of the new algorithm is that the filter coefficients are evolved in a random manner once the filter is found to be stuck at a local minimum or to have a slow convergence rate. Only the fittest coefficient set survives and is adapted according to the gradient-descent algorithm until the next evolution. As the random perturbation will be subject to the stability constraint, the filter can always minimum in a stable manner and achieve a smaller error performance with a fast rate. The article reviews adaptive IIR filtering and discusses common learning algorithms for adaptive filtering. It then presents a new learning algorithm based on the genetic search approach and shows how it can help overcome the problems associated with gradient-based and GA algorithms     

A game theoretic approach for power optimization during behavioral synthesis

In this paper, we describe a new methodology based on game theory for minimizing the average power of a circuit during scheduling and binding in behavioral synthesis. The problems are formulated as auction-based noncooperative finite games for which solutions are proposed based on the Nash equilibrium. In the scheduling algorithm, a first-price sealed-bid auction approach is used while, for the binding algorithm, each functional unit in the datapath is modeled as a player bidding for executing an operation with the estimated power consumption as the bid. Further, the techniques of functional unit sharing, path balancing, and register assignment are incorporated within the binding algorithm for power reduction. The combined scheduling and binding algorithm is formulated as a single noncooperative auction game with the functional units in the datapath modeled as players bidding for executing the operation in a particular control cycle. The proposed algorithms yield power reduction without any increase in area overhead and only a slight increase in the latency for some of the benchmark circuits. Experimental results indicate that the proposed game theoretic solution for binding yields an improvement of 13.9% over the linear programming (LP) method, while the scheduling and the combined scheduling and binding algorithms yield average improvements of 6.3% and 11.8%, respectively, over the integer-linear programming (ILP) approach.     

A fast bit-loading algorithm for high speed power line communications

Adaptive bit-loading is a key technology in high speed power line communications with the Orthogonal Frequency Division Multiplexing (OFDM) modulation technology. According to the real situation of the transmitting power spectrum limited in high speed power line communications, this paper explored the adaptive bit loading algorithm to maximize transmission bit number when transmitting power spectral density and bit error rate are not exceed upper limit. With the characteristics of the power line channel, first of all, it obtains the optimal bit loading algorithm, and then provides the improved algorithm to reduce the computational complexity. Based on the analysis and simulation, it offers a non-iterative bit allocation algorithm, and finally the simulation shows that this new algorithm can greatly reduce the computational complexity, and the actual bit allocation results close to optimal.     

An optimization approach to the frequency-domain inverse problemfor a nonuniform LCRG transmission line

The inverse problem for a nonuniform LCRG transmission line is considered in the frequency domain. Imbedding equations for the reflection and transmission coefficients are derived through the concept of wave-splitting. An optimization approach is applied to reconstruct the line parameters as functions of the position using band-limited reflection and/or transmission data. Exact and explicit expressions for the gradients are derived, and the reconstruction algorithm (based upon a conjugate gradient method) is tested with both clean and noisy data, The problem of the nonuniqueness is also discussed     

面向ScholarSpace知识库的关键词查询方法

知识库中存储着大量关于真实世界中的实体信息及实体之间的关系,随着规模的不断增长,其应用也愈发广泛。另一方面,由于大量互联网用户通过关键词描述问题和查询意图,因此如何让知识库具备更好的关键词查询应答能力,成为了研究的热点。从中文知识库的构建入手,提出了一套完整的面向中文限定领域知识库的关键词检索框架,实现并改进了基于模板的关键词查询转换方法,提出了基于语义的知识库释义和实体索引方法,提高了关键词映射能力。同时在SPARQL转换过程中采用了缺失关系索引,提高了转换成功率,提升了能够处理的查询数量。同时在学术空间ScholarSpace上对该框架进行了系统实现,取得了良好的应用效果。     

A NUMERICAL METHOD OF DIRECT SEARCH OPTIMIZATION FOR PARAMETERS OF RECIPROCAL LOSSLESS TWO-PORT NETWORK

A numerical method of direct search optimization is presented for calculating param-eters of a reciprocal lossless two-port network(RLTPN).Some examples are given.It is pointedout that the results obtained by different methods agree with one another.