Mining Adaptive Ratio Rules from Distributed Data Sources

Different from traditional association-rule mining, a new paradigm called Ratio Rule (RR) was proposed recently. Ratio rules are aimed at capturing the quantitative association knowledge, We extend this framework to mining ratio rules from distributed and dynamic data sources. This is a novel and challenging problem. The traditional techniques used for ratio rule mining is an eigen-system analysis which can often fall victim to noise. This has limited the application of ratio rule mining greatly. The distributed data sources impose additional constraints for the mining procedure to be robust in the presence of noise, because it is difficult to clean all the data sources in real time in real-world tasks. In addition, the traditional batch methods for ratio rule mining cannot cope with dynamic data. In this paper, we propose an integrated method to mining ratio rules from distributed and changing data sources, by first mining the ratio rules from each data source separately through a novel robust and adaptive one-pass algorithm (which is called Robust and Adaptive Ratio Rule (RARR)), and then integrating the rules of each data source in a simple probabilistic model. In this way, we can acquire the global rules from all the local information sources adaptively. We show that the RARR technique can converge to a fixed point and is robust as well. Moreover, the integration of rules is efficient and effective. Both theoretical analysis and experiments illustrate that the performance of RARR and the proposed information integration procedure is satisfactory for the purpose of discovering latent associations in distributed dynamic data source.     

Attribute selection based on information gain ratio in fuzzy rough set theory with application to tumor classification

Tumor classification based on gene expression levels is important for tumor diagnosis. Since tumor data in gene expression contain thousands of attributes, attribute selection for tumor data in gene expression becomes a key point for tumor classification. Inspired by the concept of gain ratio in decision tree theory, an attribute selection method based on fuzzy gain ratio under the framework of fuzzy rough set theory is proposed. The approach is compared to several other approaches on three real world tumor data sets in gene expression. Results show that the proposed method is effective. This work may supply an optional strategy for dealing with tumor data in gene expression or other applications.     

Simulating Natural Light

Simulating natural light is an interesting subject in computer graphics. Using the principles of meteorology, we build a sphere cap light source model . Unlike existing illumination or light source models, we consider the effects of both the ratio of direct light to diffuse light and the position of the sun under different weather conditions. We use an area light source to generate shadows and a point light source to generate the other effects. The ratio of the two types of light is adjusted by a brightness factor b while the position of the sun is determined by giving the light direction. In addition. a modifiable sphere cap is used to obtain modified penumbra which vary with illumination.     

基于FPGA的TFT LCD快检信号源的设计

目前TFT模块的生产过程中常伴有模块的线缺陷和点缺陷,依据TFF模块的驱动和测试原理,设计了一种由FPGA和模拟开关组成的集成测试信号源,该信号源可提供源极信号、栅极信号、栅极控制信号和公共地信号四路测试信号,此四路信号无论是在频率、占空比、延时,还是幅值上都满足对一般中小尺寸TFT液晶模块的线缺陷和点缺陷的测试要求,可以灵活方便地移植到其他模块测试中,具有通用性.     

A null space method for over-complete blind source separation

In blind source separation, there are M sources that produce sounds independently and continuously over time. These sounds are then recorded by m receivers. The sound recorded by each receiver at each time point is a linear superposition of the sounds produced by the M sources at the same time point. The problem of blind source separation is to recover the sounds of the sources from the sounds recorded by the receivers, without knowledge of the m×M mixing matrix that transforms the sounds of the sources to the sounds of the receivers at each time point. Over-complete separation refers to the situation where the number of sources M is greater than the number of receivers m, so that the source sounds cannot be uniquely solved from the receiver sounds even if the mixing matrix is known. In this paper, we propose a null space representation for the over-complete blind source separation problem. This representation explicitly identifies the solution space of the source sounds in terms of the null space of the mixing matrix using singular value decomposition. Under this representation, the problem can be posed in the framework of Bayesian latent variable model, where the mixing matrix and the source sounds can be inferred based on their posterior distributions. We then propose a null space algorithm for Markov chain Monte Carlo posterior sampling. We illustrate the algorithm using several examples under two different statistical assumptions about the independent source sounds. The blind source separation problem is mathematically equivalent to the independent component analysis problem. So our method can be equally applied to over-complete independent component analysis for unsupervised learning of high-dimensional data.     

基于极值约束的鞋底边缘立体匹配

为了提取鞋底边缘特征,提出了一种中间加光源的双目立体视觉测量方法,增大边缘内外的对比度。针对鞋头和鞋跟特征匹配,提出了基于极值约束的边缘立体匹配方法。该方法分三步：采用Canny算子提取左右图像的边缘;对鞋跟和鞋头部分的边缘点利用最小二乘三项式曲线拟合,求取边缘点上与曲线极值点最近的点作为极值特征点;利用长度均分法进行其他边缘点的匹配。对鞋底进行双目三维测量实验,结果表明,中间加光源可使得图像边缘清晰,便于边缘提取;提出的匹配方法获得的鞋底边缘三维数据完整、正确,有效地抑制了匹配错误。     

基于多频段排序的语音卷积混合盲源分离

针对语音卷积盲源分离频域法排列顺序不确定性问题,提出一种多频段能量排序算法。首先,通过对混合信号的短时傅立叶变换(STFT),在频域上各个频点建立一个瞬时混合模型进行独立分量分析,之后结合能量相关排序法和波达方向(DOA)排序法解决排序不确定性问题,再利用分裂语谱方法解决幅度不确定性问题,进而得到每个频点正确的分离子信号,最后利用逆短时傅立叶(ISTFT)变换得到分离的源信号。仿真结果表明,与Murata的排序算法对比,改进的算法在信号偏差比、信道干扰比、系统误差比上都所提高。     

基于垂线的WSN基站位置隐私保护算法

针对无线传感器网络基站位置隐私保护问题,提出一种基于垂线的基站位置隐私保护算法。根据源节点的位置与坐标轴上一点随机确定一条直线,过基站做直线的垂线,源节点关于该垂线对称的点为第1个预期幻影源节点,使第1个幻影源节点分布具有地理位置多样性。以第1个预期幻影源节点为垂足建立垂线,在该垂线上确定第2个预期幻影源节点,为第2个幻影源节点提供选择方向,2个幻影源节点不仅能够为数据包传输提供多样性,而且分别沿着直线和垂线传输假包,诱导攻击者向远离基站的方向追踪,增大攻击者捕获基站的难度。仿真结果表明,该算法能够有效地诱导攻击者偏离真实路径,提高安全周期。     

求带多个限制条件的单源多权最短路径算法

带限制条件的多权最短路径问题具有广泛的应用。本文给出一个通过按字典序生成从源顶点到目标顶点的非支配路径的方法,求出满足限制条件的最短路径的算法,并且分析了算法的时间复杂度。     

基于视频的大空间建筑消防系统研究

针对基于视频的大空间建筑火灾消防存在的实时性和有效性问题,提出了一种基于改进的自适应差分进化算法的摄像机自标定方法。用基于比值法与相关函数法融合的SURF火灾图像特征点匹配算法,快速得到准确的匹配点对;进行摄像机自标定,用较为准确的匹配点对求得基本矩阵F,利用绝对二次曲线的性质,得到优化函数。利用基于改进的差分进化算法对其进行优化,求得摄像机内参数,得到火源的三维信息。实验结果证明,该方法短时间内,算出了较为准确的火源的空间三维信息,实时性和精确度均能够满足火灾消防的标准,有效地进行灭火。     

基于互相关时间延迟估计的心音源定位方法

提出一种用于正常噪声环境下的心音源定位方法。利用3个传感器同步记录胸前3个位置的心音信号,对自动识别出的心音成分采用互相关法,估计同一心音源发出的信号到达2个传感器的时间延迟。利用时间延迟和传感器坐标确定心音源在传感器连线上的投影点坐标,由投影点与心音源的关系定位心音源。实验结果显示,对于采集到的健康人心音信号,利用该方法可以定位第一心音源。     

基于麦克风阵列的声源定位系统设计

从系统的角度对真实声场环境下基于到达时间差的声源定位算法进行了研究,搭建了基于麦克风阵列的声源定位系统,将LabVIEW应用于声源定位软件的开发,方便地完成信号的采集与处理,运用广义互相关函数法进行时延估计,结合平面四元十字阵模型建立方程组实现定位。该系统人机界面友好,具有较强的数据分析和处理能力,能及时跟踪声源的位置变化。     

以优先点为中心的Delaunay三角网生长算法

目的 Delaunay三角网具备的优良性质使其得到广泛的应用,构建Delaunay三角网是计算几何的基础问题之一,为了高效、准确地构建大规模点集的Delaunay三角网,提出一种基于优先点的改进三角网生长算法.方法 算法以逆时针次序的一条凸包边为初始基边,使用基边对角最大化并按照逆时针次序选定第3点构建一个Delaunay三角形,通过待扩展边列表中的数据判断新生成的两条边是否需要扩展,采用先进先出的方式从待扩展边列表中取边作为基边,以优先点为中心构建局部Delaunay三角网使优先点尽快成为封闭点,再从点集中删除此封闭点.结果 对于同一测试点集,改进算法运行时间与经典算法运行时间的比率不超过1/3,且此比率随点集规模增长逐步下降.相比经典算法,改进算法在时间效率上有较大提升.结论 本文改进算法对点集规模具有较好的自适应性与较高的构网效率,可用于大规模场景下Delaunay三角网的构建.     

Original source tracing enabled by e-learning contents system based on crowdsourcing

A crowdsourcing environment, where there is a very large volume of diverse content resulting from the participation of a mass of unspecified individuals, has resulted in significant changes in education. This paper presents an e-learning content system to manage the inclusion of crowdsourced material on the Web within lecture materials. The e-learning content system comprises a scrape system, learning content editor, and tracing system. As Web content may change with the progress of time, teachers (and students) must check whether the Web-based materials previously used in their classes have been updated. Accordingly, we designed scrape metadata specifications for tracing the original source. These metadata include information on copyrights and tracing, rather than basic data regarding the original source, to allow users to determine whether the original source has been updated. An editor was also configured so that the scraped Web content could be immediately incorporated into the teaching materials for enhanced convenience. The change point tracing accuracy test and utility evaluation performed using this system show that the accuracy of the change point tracing was 97.1% and that this system effectively saves time as compared with checking for changes by entering each URL directly.     

基于相关系数的适用于视频拼接的帧对齐方法

为解决视频拼接中源视频在时间上不同步的问题,提出了一种视频帧对齐方法.利用图像矩阵间的相关系数特性,根据视频拼接中源视频的特点,运用运动量检测方法检测视频中不同时间点的运动量,在低运动量点运用相位相关法确定交叠区域大致范围,在高运动量点计算相关系数确定同一时间点上的对应帧图像.在对应帧的搜索过程中,使用变步长搜索算法提高搜索效率,减少搜索时间.实验中使用无标定的双摄像机采集多组视频数据,结果表明了该方法的准确性和有效性.     

Planning constrained motion

We consider the motion planning problem for a point constrained to move along a path with radius of curvature at least one. The point moves in a two-dimensional universe with polygonal obstacles. We show the decidability of the reachability question: “Given a source placement (position and direction pair) and a target placement, is there a curvature-constrained path from source to target avoiding obstacles?” The decision procedure has time and space complexity 2 ^{o(poly(n, m))} wheren is the number of corners andm is the number of bits required to specify the position of corners.     

基于量子遗传算法的高效匹配搜索策略

在大规模源图像上进行图像匹配时,最佳匹配点的搜索策略是匹配算法时间性能的决定因素,设计高效匹配搜索策略是提高算法性能的关键。为了减少搜索时间和提高匹配实时性,本文基于匹配源图像划分和量子遗传算法基本原理,提出了面向大规模源图像匹配的目标淘汰搜索策略TESS。TESS将基于整幅源图像的全空间随机搜索的过程变成基于各个子图像的子空间并行搜索和逐步淘汰的过程,实现了匹配区域粗定位与匹配点精搜索的有效结合,从而大大缩短了最佳匹配点的搜索时间。实验结果表明,TESS搜索策略带来了匹配速度的极大提高,且时间加速比随匹配源图像规模的增大而增大。     

Closed-form solution to motion parameter estimation of an acoustic source exploiting Doppler effect

This paper proposes an analytical solution for the parameter estimation of a moving acoustic source from the instantaneous frequency of a recorded signal on a ground located microphone. The acoustic source could be either a propeller-driven aircraft or a helicopter. During the transit of such acoustic source, the Doppler effect is exploited to estimate the speed, altitude, source frequency and the closest point of approach (CPA) time between the source and the microphone. The proposed solution is an original workout of an existing time–frequency based approach which has been applied previously to localize a propeller-driven aircraft. In contrast to existing solutions, the main advantage of the proposed one is the direct estimation of the flight parameters without resorting to any iterative procedure. The proposed approach is faster, less complex and therefore a good candidate for hardware implementation. The proposed solution is compared with respect to the existing approach in term of parameter estimation, performance and computation complexity. Experimental validations of the proposed solution are also presented.     

A target tracker using spatially distributed infrared measurements

An extended Kalman filter algorithm is designed to track a point source target in an open-loop tracking problem, using outputs from a forward-looking infrared (FLIR) sensor as measurements. The filter separately estimates the translational position changes of the target in the FLIR field of view due to two effects: actual target motion and apparent motion caused by atmospheric turbulence. A Monte Carlo analysis is conducted to determine the performance of the filter as a function of signal-to-noise ratio, target spot size, the ratio of rms target motion to rms atmospheric jitter, target correlation times, and mismatches between the true target spot size and the size assumed by the filter. The performance of the extended Kalman filter is compared to the performance of an existing correlation tracker under identical conditions. A one sigma tracking error of 0.2 and 0.8 picture elements is obtained with signal-to-noise ratios of 20:1 and 1:1, respectively. No degradation in performance is observed when the spot size is decreased or when the target correlation time is increased over a limited range, when filter parameters are adjusted to reflect this knowledge. Sensitivity analysis shows that the filter is robust to minor changes in target intensity spot size.     

Allow plenty of time for large-scale software

Estimation of development time for multimillion-line software is addressed. Three methods are discussed. The historical-database approach finds the statistical trend line by regression analysis that relates project development time to size in source lines, and extends it to a ten-million-line size. The software macroestimation methods, Cocomo (cost constructive model) and the software life-cycle model, forecast labor effort and schedule from early estimates of the amount of functionality in the project. The three methods all point to the conclusion that a large amount of time must be allowed