收敛性随机森林模型及其遥感应用

通过引入全局损失函数,提出了一种全局优化的随机森林模型算法,称为θ-β型随机森林,并且利用改进后的模型对城市遥感图进行了检测与识别,识别准确率与识别速率都得到了一定的提高.方法在经典随机森林模型的基础上加入前向反馈模型(Forward Stagewise Additive Model),通过每一层节点的训练结果干预下一层的训练数据(从而改变阈值θ的选择)与训练步长(β),使得最后训练得到的型随机森林收敛速度更快,预测结果更为准确.     

混沌遗传算法对BP神经网络的改进研究

BP神经网络算法是目前应用最广泛的一种神经网络算法,但有收敛速度慢和易陷入局部极小值等缺陷.本文利用混沌遗传算法(CGA)具有混沌运动遍历性、遗传算法反演性的特性来改进BP神经网络算法.该算法的基本思想是用混沌遗传算法对BP神经网络算法的初始权值和初始阈值进行优化.把混沌变量加入遗传算法中,提高遗传算法的全局搜索能力和收敛速度;用混沌遗传算法优化后得到的最优解作为BP神经网络算法的初始权值和阈值.通过实验观察,改进后的结果与普通的BP神经网络算法的结果相比,具有更高的准确率.     

正交变换在图像压缩中的应用

用矩阵表示图像,构造正交均值差分变换矩阵,对原始图像进行正交变换,进一步取阈值,仅存储绝对值大于阈值的系数,获得数据压缩.解压缩过程只需作逆均值差分变换.最后将该算法分别应用于灰度和彩色图像的压缩处理,结果验证了算法的有效性.由于算法中所有变换都通过矩阵运算处理,且意义直观明了,故该算法是大学线性代数教学中一个非常好的应用案例.     

基于改进阈值的马田系统在医疗智能诊断中的应用

人工智能与医疗数据的融合可以加速疾病诊断过程,提高诊断精度,挖掘诊断过程关键指标,改善医疗工作流程.以医疗决策者偏好为前提,提出基于二次损失函数改进阈值的马田系统算法.通过正交表与信噪比对指标进行优化,降低模型复杂度;通过对阈值的改进,提高模型的灵敏度,满足医疗工作者的决策偏好.将此方法应用于UCI乳腺癌医学数据及三甲医院哮喘临床数据,并与其他改进阈值的马田系统算法及智能算法对比分析.结果表明,改进阈值的马田系统算法识别灵敏度高,简化诊断指标,训练耗时少,是一种更为有效的医疗智能诊断方法.     

矩阵填充的硬阈值修正算法

提出了使用硬阈值进行矩阵填充的修正算法.算法通过对迭代矩阵进行对角修正来完成矩阵填充,并给出了算法的收敛性分析.最后通过数值实验比较了修正算法与硬阈值算法填充的数值结果,显示出了新算法的优越性.     

识别含热源瞬态热传导问题的热扩散系数

针对含有热源的瞬态热传导反问题,引入一个变换将含热源热传导问题转换为无热源热传导问题,采用改进布谷鸟算法反演热扩散系数.正问题由边界元法求解.将热扩散系数作为优化变量,以计算温度和测量温度之间的接近程度为目标函数,通过改进布谷鸟算法极小化目标函数来优化估计热扩散系数.比较共轭梯度法、布谷鸟算法和改进布谷鸟算法的反演结果.与共轭梯度法相比,改进布谷鸟算法对迭代初值不敏感;与布谷鸟算法相比,改进布谷鸟算法收敛速度更快.算例讨论了测点数量、鸟巢数量、测量误差对计算结果的影响.增加测点数量,反演结果精度降低;增加鸟巢数量,迭代次数减少;随着测量误差的增大,结果精度降低.数值算例验证了改进布谷鸟算法反演热扩散系数的准确性和有效性.     

基采用时频区域检测的雷达信号盲分选算法

为了解决欠定条件下密集雷达信号分选问题,提高雷达信号盲分选算法精度,提出了一种采用时频区域检测的雷达信号盲分选算法,首先利用短时傅立叶变换将混合信号映射至时频平面进行处理,然后通过聚类算法估计出混合矩阵,从而反解出源信号矩阵,进而估计出每一雷达源信号,能够在信号的时频平面投影相交且欠定的条件下,实现信号分选功能.仿真实验结果表明:提出的算法相比于传统的类MUSIC算法及其衍生的相关改进算法具有更高的分选精度和算法收敛性,且估计得到的源信号时域波形更优,体现了其有效性和优越性.     

基于PSO_Trainlm BP模型的图像去噪研究

针对在使用BP模型进行图像去噪时,模型存在的对初始权阈值敏感、易陷入局部极小值和收敛速度慢的问题.为了提高模型去噪效率,提出采用改进粒子群神经网络模型进行图像去噪.首先运用改进粒子群算法对BP神经网络权阈值进行初始寻优,再用trainlm BP算法对优化的网络权阈值进一步精确优化,随后建立基于粒子群算法的BP神经网络去噪模型,并将其应用到图像去噪研究中.仿真结果表明,新模型结合了粒子群算法的全局寻优能力和BP算法的局部搜索能力,减小了模型对初始权阈值的敏感性,有效防止了模型陷入局部极小值的可能,提高了图像去噪模型的速度和质量.     

基于模拟退火算法的城市物流多目标配送车辆路径优化研究

基于传统模拟退火算法,通过引入记忆函数、结合GIS手段,运用SPSS聚类分析来确定初始化状态种群、多种群并行机制和新状态的产生.依据种群规模采用不同产生算法来改进算法,并将改进算法应用于城市物流中确定的多目标车辆路径优化问题,验证了算法的可行性与实用价值.此外,改进算法分别与传统模拟退火算法和GIS图解法相比较,优化效率和准确率都得到了很大的提高.     

圆内两随机点距离的概率分布

研究了圆内两个均匀分布随机点之间距离的概率分布.利用几何概率推导出了概率分布函数、概率密度函数和数学期望的解析表达式.基于R语言,对单位圆内随机点之间距离进行了随机模拟,模拟结果与理论分析结果互为印证.     

Geometric Entities Voting Schemes in the Conformal Geometric Algebra Framework

Traditional methods for geometric entities resort to the Hough transform and tensor voting schemes for detect lines and circles. In this work, the authors extend these approaches using representations in terms of k-vectors of the Conformal Geometric Algebra. Of interest is the detection of lines and circles in images, and planes, circles, and spheres in the 3-D visual space; for that, we use the randomized Hough transform, and by means of k-blades we code such geometric entities. Motivated by tensor voting, we have generalized this approach for any kind of geometric entities or geometric flags formulating the perceptual saliency function involving k-vectors. The experiments using real images show the performance of the algorithms.     

Detecting Curves of Symmetry in Images Via Hough Transform

The Hough transform is a standard pattern recognition technique introduced between the 1960s and the 1970s for the detection of straight lines, circles, and ellipses with several applications including the detection of symmetries in images. Recently, based on algebraic geometry arguments, the procedure has been extended to the automated recognition of special classes of algebraic plane curves. This allows us to detect curves of symmetry present in images, that is, curves that recognize midpoints maps of various shapes extracted by an ad hoc symmetry algorithm, here proposed. Further, in the case of straight lines, the detection of lines of symmetry allows us, by a pre-processing step of the image, to improve the efficiency of the recognition algorithm on which the Hough transform technique is founded, without loss of generality and additional computational costs.     

Statistical properties of the Hough transform estimator in the presence of measurement errors

The Hough transform is a common computer vision algorithm used to detect shapes in a noisy image. Originally the Hough transform was proposed as a technique for detection of straight lines in images. In this paper we study the statistical properties of the Hough transform estimator in the presence of measurement errors. We consider the simple case of detection of one line parameterized in polar coordinates. We show that the estimator is consistent, and possesses a rate of convergence of the cube-root type. We derive its limiting distribution, and study its robustness properties. Numerical results are discussed as well. In particular, based on extensive experiments, we define a “rule of thumb” for the determination of the optimal width parameter of the template used in the algorithm.     

基于特殊混料模型D-最优设计搜索的交换点式门限接受算法

在搜索混料模型D-最优设计的计算机算法领域,主流算法包括经典的Fedorov算法,以及元启发类算法,但两者在一些特定的优化问题上,分别在收敛速度和收敛精度方面有进一步提升的空间.文章分别探讨了可能造成这种情况的两类算法各自的局限性,并采取优势互补的策略,构建了交换点式门限接受算法,即ETA (exchange threshold accepting)算法.以含倒数项混料模型为例,文章验证了ETA算法生成设计的D-最优性,并分别与Fedorov算法和元启发类的ProjPSO算法作比较.结果表明,至少在某些特殊的混料模型D-最优设计的搜索方面,ETA算法在收敛速度和精度方面均具有一定的优势.     

求解等圆Packing问题的完全拟物算法

沿着拟物的思路进一步研究了具有NP难度的等圆Packing问题.提出了两个拟物策略,第一个是拟物下降算法,第二是让诸圆饼在某种物理定律下做剧烈运动.结合这两个策略,提出了一个统一的拟物算法.当使用N(N=1,2,3,…,100)等圆最紧布局的国际记录对此算法进行检验时,发现对于N=66,67,70,71,77,89这6个算例,本算法找到了比当前国际纪录更优的布局.     

Solution Methodologies for the Smallest Enclosing Circle Problem

Given a set of circles C = {c ₁, ..., c _n} on the Euclidean plane with centers {(a ₁, b ₁), ..., (a _n, b _n)} and radii {r ₁, ..., r _n}, the smallest enclosing circle (of fixed circles) problem is to find the circle of minimum radius that encloses all circles in C. We survey four known approaches for this problem, including a second order cone reformulation, a subgradient approach, a quadratic programming scheme, and a randomized incremental algorithm. For the last algorithm we also give some implementation details. It turns out the quadratic programming scheme outperforms the other three in our computational experiment.     

Optimal Estimator for Distributed Anonymous Observers

In this paper, we consider a distributed estimation problem in which multiple observations of a signal process are combined via the maximum function for the decision making. A key result established is that, under suitable technical conditions, the optimal decision function can be implemented by means of thresholds. A natural question is how to determine the optimal threshold value. We propose here an algorithm for threshold adjustment by means of training sequences. The algorithm is a variation of the Kiefer-Wolfowitz algorithm with expanding truncations and randomized differences. A result of the paper is to establish the convergence of the algorithm if the variance of observation noises is small enough.     

Track initiation with ant colony optimization

Since the issue of track initiation belongs to the NP-hard problem in the bearings-only multi-sensor-multi-target tracking system, a novel proposed track initiation technique is proposed in this paper. The proposed track initiation technique is based upon an ant colony optimization (ACO) algorithm, a kind of heuristic optimization method. Observing that each target is of the characteristic of uniform rectilinear motion, we develop a new cost function derived from the thought of Hough transform. Numerical simulation results show that the proposed ACO-based track initiation method not only meets the requirement of real time, but also performs better than other traditional techniques, especially in the scenario that all targets move in parallel.     

基于噪声先验知识的LMS自适应噪声对消法的水声微弱信号提取

讲述了LMS自适应噪声对消法的数学原理,设计了一种基于噪声参考信号的噪声对消原理结构图,并对初步提取的信号做Fourier变换,设计合理的系数阈值经行滤波,Fourier逆变换的信号与理想信号做性能对比.仿真实验表明:基于噪声参考信号的噪声对消算法呈现出滤波阶数少,收敛速度快,精度良好,提取信号效果良好等优点.汾河二库湖试测试实验结果也验证了该算法具有良好的高效性和实用性.