遗传模糊聚类算法在图像边缘检测中的应用

将一种改进的遗传模糊c-均值聚类(GFGA)算法应用到图像的边缘检测中.我们将灰度图像中的每一个像素点看成是一个数据样本,将该点的灰度值经过Robert算子、Sobel算子和Prewitt算子处理构成它的特性向量,形成具有三维特征的数据集,然后对这个数据集应用遗传模糊聚类算法进行分类,自适应地检测出图像的边缘点,达到提取边缘的目的.实验结果表明,这种混合算法能得到很好的边缘效果,并且得到的结果无需再细化处理,提高了边缘定位的精度.     

基于多特征与卷积神经网络的SAR图像分类方法

针对合成孔径雷达图像的分类优化方法,提出一种基于多特征与卷积神经网络的SAR图像分类方法Canny-WTD-CNN.将Canny算子提取的边缘特征,与小波阈值去噪法提取的小波特征进行自适应融合,作为卷积神经网络的输入;以softmax为分类器,对SAR图像进行分类识别检测.最后利用MSTAR公开数据集的三类目标数据进行试验,并给出该方法与其他方法结果的对比,表明该方法的有效性,识别率达到99.14%.     

基于人脸重要特征和SVM的人脸识别方法

提出了一种基于人脸重要特征的人脸识别方法,首先选取人脸的重要特征并将其具体化,对得到的重要特征进行主成分分析,然后用支持向量机(Support Vector Machine,SVM)设计重要特征分类器来确定测试人脸图像中重要特征,同时设计支持向量机(SVM)人脸分类器,确定人脸图像的所属类别.对ORL人脸图像数据库进行仿真实验,结果表明,该方法要优于一般的基于整体特征的人脸识别方法并有较强的鲁棒性.     

基于区域推荐和深度卷积网络的交通目标检测

为解决传统目标检测算法需要人工设定目标特征、使用滑动窗法判断目标可能区域耗时等问题,将基于区域推荐和深度卷积网络用于交通目标检测,直接从原始图像提取特征,免去了人工选取特征的环节;解决了滑动窗口法耗时的问题.首先采用Selective Search方法在源图像上生成大量的候选区域,以这些候选区域作为输入样本,训练深度卷积网络学习算法,自动进行特征提取,对每个候选区提取的特征采用SVM分类器进行分类,最后基于贪婪非极大值抑制方法精修候选框的位置.此算法通过matlab编程分别对单目标、多目标及多类交通目标进行检测实验,证明了所提方法的可行性和有效性.     

基于LSD的具有特定特征的三维物体轮廓提取算法

提出一种基于LSD的具有特定特征的三维物体轮廓提取算法,该算法主要用于检测结构环境中的特定目标,提取其轮廓目的在于计算出目标的位姿信息,便于机器人作业抓取.针对具有特定特征的三维物体,算法首先对物体进行基于LSD的外轮廓提取,其次再对原始图像进行基于改进的Prewitt梯度算子的边缘提取,并删除经边缘提取后外轮廓特征信息,再进行基于LSD的物体内轮廓提取,从而实现物体的轮廓提取.基于LSD的轮廓提取算法包括提取梯度图像,LSD线段检测,直线段的分类和轮廓的拟合四个步骤.在直线段分类中,提出了利用极坐标表示线段特征,利用极径和极角对线段进行有效分类的方法.为了进一步证明三维轮廓提取算法的有效性,以长方体为例,详细地阐述了该算法的具体步骤.     

改进Canny算法在钢印数字特征提取中的应用

边缘检测是实现图像分割、特征提取和图像理解的基础.研究了传统Canny算子的优势与不足.在此基础上,提出了一种快速分块自适应Canny算法.方法首先按字符大小分割图像,然后在每一块上进行自适应边缘检测.自适应边缘检测是在平滑图像的同时得到高斯滤波尺度参数,然后采用Otsu方法的自适应阈值计算Canny算子的高、低门限值.实验结果表明,方法不需人工设定参数就能自动提取不同光照背景下的钢印数字边缘,而且能有效抑制噪声,与传统Canny算子相比,边缘连接程度最佳,噪声敏感程度较低,实时性较强.     

基于概率弹性匹配的多姿态人脸认证

为解决多姿态情况下的人脸认证问题,提出了基于概率弹性匹配的方法.首先从多尺度密集采样的图像中提取局部特征,并将局部区域的位置信息保存到特征向量中.然后用高斯混合模型在图像集中训练得到人脸特征的空间外观分布,并将人脸图像中的每一个局部区域表示成与其最相似的高斯分量,从而形成一个基于高斯混合模型表示的人脸特征.通过统计两个人脸特征向量的差向量来训练支撑向量机分类器从而实现多姿态人脸认证.在通用图像库中的实验验证了方法的有效性.     

基于改进Gentle-Adaboost算法的疲劳驾驶中人脸检测研究

疲劳驾驶检测中人脸检测的算法有很多,其中相对成熟的是Gentle-Adaboost,此算法虽然识别率较高,但是识别时间较长,研究在识别过程中使用均值哈希算法进行算法优化,并引入缓存的概念来缩短识别时间.首先,根据Gentle-Adaboost算法和Haar特征来提取基本特征点,然后通过对比前后两帧图像的哈希指纹来减少人脸检测次数,最后通过缓存数据库来存储相似图片的哈希指纹,运行一段时间后仅需通过对比哈希指纹就能精准的找到人脸区域.通过实验,可以证明改进后的平均识别时间可减少原时间的80%.     

一种结合小波模极大值法和Canny算子的目标提取方法

针对基于小波变换的目标提取中忽略低频子图像的一些重要信息的问题.提出了一种基于小波变换的模极大值法和Canny算子的目标提取方法.在小波域中,通过求解局部小波系数模型的极大值点提取(检测)高频边缘,利用Canny算子提取(检测)低频边缘.然后根据融合规则对两个子图像边缘进行融合.实验结果表明,该方法不仅能有效地增强图像边缘,而且能准确地定位图像边缘.     

基于随机谱聚类的图像分割算法

图像分割就是把感兴趣的区域从背景中分割、提取出来,为了使分割出来的图像特征信息完整,根据图像的灰度值和空间距离构造了一种相似度函数,得到基于图的灰度值的相似度矩阵,将图像分割转化为图论最小割问题,然后运用谱聚类算法进行分割.针对谱聚类算法运行所需的内存空间和运算量大的特点,提出一种考虑概率因素的随机抽样谱聚类算法.在具体实施时,为了减少背景噪声对分割结果的影响,对图像进行了滤波预处理.结果表明,算法稳定性好,相对现有算法,分割效果得到改善.     

Weighted fusion frame construction via spectral tetris

Fusion frames consist of a sequence of subspaces from a Hilbert space and corresponding positive weights so that the sum of weighted orthogonal projections onto these subspaces is an invertible operator on the space. Given a spectrum for a desired fusion frame operator and dimensions for subspaces, one existing method for creating unit-weight fusion frames with these properties is the flexible and elementary procedure known as spectral tetris. Despite the extensive literature on fusion frames, until now there has been no construction of fusion frames with prescribed weights. In this paper we use spectral tetris to construct more general, arbitrarily weighted fusion frames. Moreover, we provide necessary and sufficient conditions for when a desired fusion frame can be constructed via spectral tetris.     

A skin detection approach based on the Dempster–Shafer theory of evidence

Skin detection is an important step for a wide range of research related to computer vision and image processing and several methods have already been proposed to solve this problem. However, most of these methods suffer from accuracy and reliability problems when they are applied to a variety of images obtained under different conditions. Performance degrades further when fewer training data are available. Besides these issues, some methods require long training times and a significant amount of parameter tuning. Furthermore, most state-of-the-art methods incorporate one or more thresholds, and it is difficult to determine accurate threshold settings to obtain desirable performance. These problems arise mostly because the available training data for skin detection are imprecise and incomplete, which leads to uncertainty in classification. This requires a robust fusion framework to combine available information sources with some degree of certainty. This paper addresses these issues by proposing a fusion-based method termed Dempster–Shafer-based Skin Detection (DSSD). This method uses six prominent skin detection criteria as sources of information (SoI), quantifies their reliabilities (confidences), and then combines their confidences based on the Dempster–Shafer Theory (DST) of evidence. We use the DST as it offers a powerful and flexible framework for representing and handling uncertainties in available information and thus helps to overcome the limitations of the current state-of-the-art methods. We have verified this method on a large dataset containing a variety of images, and achieved a 90.17% correct detection rate (CDR). We also demonstrate how DSSD can be used when very little training data are available, achieving a CDR as high as 87.47% while the best result achieved by a Bayesian classifier is only 68.81% on the same dataset. Finally, a generalized DSSD (GDSSD) is proposed achieving 91.12% CDR.     

基于最优估计的数据融合理论

本文提出了一种最优加权的数据融合方法，分析了最优权值的分配原则；给出了多源信息统一的线性融合模型，使其表示不受数据类型和融合系统结构的限制，并指出在噪声协方差阵正定的前提下，线性最小方差估计融合和加权最小二乘估计融合是等价的；介绍了数据融合中的Bayes极大后验估计融合方法，给出了利用极大后验法进行传感器数据融合的一般表示公式；最后以两传感器数据融合为例，证明了利用Bayes极大后验估计进行两传感器数据融合所得到的融合状态的精度比相同条件下极大似然估计得到的精度要高，同时它们均优于任一单传感器局部估计精度。     

Volume growth and spectrum for general graph Laplacians

We prove estimates relating exponential or sub-exponential volume growth of weighted graphs to the bottom of the essential spectrum for general graph Laplacians. The volume growth is computed with respect to a metric adapted to the Laplacian, and use of these metrics produces better results than those obtained from consideration of the graph metric only. Conditions for absence of the essential spectrum are also discussed. These estimates are shown to be optimal or near-optimal in certain settings and apply even if the Laplacian under consideration is an unbounded operator.     

Representing polyhedra: faces are better than vertices

This paper investigates the reconstruction of planar-faced polyhedra given their spherical dual representation. The spherical dual representation for any genus 0 polyhedron is shown to be unambiguous and to be uniquely reconstructible in polynomial time. It is also shown that when the degree of the spherical dual representation is at most four, the representation is unambiguous for polyhedra of any genus. The first result extends, in the case of planar-faced polyhedra, the well known result that a vertex or face connectivity graph represents a polyhedron unambiguously when the graph is triconnected and planar. The second result shows that when each face of a polyhedron of arbitrary genus has at most four edges, the polyhedron can be reconstructed uniquely. This extends the previous result that a polyhedron can be uniquely reconstructed when each face of the polyhedron is triangular. As a consequence of this result, faces are a more powerful representation than vertices for polyhedra whose faces have three or four edges. A result of the reconstruction algorithm is that high level features of the polyhedron are naturally extracted. Both results explicitly use the fact that the faces of the polyhedron are planar. It is conjectured that the spherical dual representation is unambiguous for polyhedra of any genus.     

基于TFWPA算子的模糊优化组合预测模型及其应用

传统的组合预测中,预测对象往往是实数或区间数,实际上,三角模糊数则更能刻画不确定环境下复杂事物的某些量的特征。因此,本文提出一种预测信息为三角模糊数的模糊优化组合预测新方法。定义了两个三角模糊数的相对误差,同时考虑到预测数据之间的交叉影响,基于三角模糊加权Power平均(TFWPA)算子、三角模糊加权Power几何(TFWPG)算子和max-min准则,分别构建模糊优化组合预测模型。提出非劣性和优性组合预测的概念,证明所提模糊组合预测模型具有非劣性质。最后通过实例分析说明了该模糊组合预测方法的有效性,并对参数做了灵敏度分析。     

Weighted approximation of functions by Szász--Mirakyan-type operators

Summary We give error estimates for the weighted approximation of functions with singularities at the endpoints on the semiaxis by some modifications of Sz\'asz--Mirakyan operators. To do so, we define a new weighted modulus of smoothness and prove its equivalence to the weighted K-functional. Also, the class of functions for which the modified Sz\'asz--Mirakyan operator can be defined will be extended to a much wider set than for the original operator.     

Small Grid Embeddings of 3-Polytopes

We introduce an algorithm that embeds a given 3-connected planar graph as a convex 3-polytope with integer coordinates. The size of the coordinates is bounded by O(2^7.55n )=O(188 ⁿ ). If the graph contains a triangle we can bound the integer coordinates by O(2^4.82n ). If the graph contains a quadrilateral we can bound the integer coordinates by O(2^5.46n ). The crucial part of the algorithm is to find a convex plane embedding whose edges can be weighted such that the sum of the weighted edges, seen as vectors, cancel at every point. It is well known that this can be guaranteed for the interior vertices by applying a technique of Tutte. We show how to extend Tutte’s ideas to construct a plane embedding where the weighted vector sums cancel also on the vertices of the boundary face.     

Fusion of multi-spectral and panchromatic images using fuzzy rule

A fusion approach is proposed to refine the resolution of a multi-spectral image using a high-resolution panchromatic image. After the two images are decomposed by wavelet transform, five texture features are extracted from the high-frequency detailed sub-images. Then a nonlinear fusion rule, i.e. fuzzy rule is used to merge wavelet coefficients from the two images according to the extracted features. Experimental results indicate that the method outperforms the traditional approaches in preserving spectral information while improving spatial information.     

Self-Adjoint Extensions of Discrete Magnetic Schrödinger Operators

Using the concept of an intrinsic metric on a locally finite weighted graph, we give sufficient conditions for the magnetic Schrödinger operator to be essentially self-adjoint. The present paper is an extension of some recent results proven in the context of graphs of bounded degree.