最小化类内距离和分类算法

支持向量机分类算法引入惩罚因子来调节过拟合和线性不可分时无解的问题,优点是可以通过调节参数取得最优解,但带来的问题是允许一部分样本错分。错分的样本在分类间隔之间失去了约束,导致两类交界处样本杂乱分布,并且增加了训练的负担。为了解决上述问题,该文根据大间隔分类思想,基于类内紧密类间松散的原则,提出一种新的分类算法,称之为最小化类内距离和(Intraclass-Distance-Sum-Minimization, IDSM)分类算法。该算法根据最小化类内距离和准则构造训练模型,通过解析法求解得到最佳的映射法则,进而利用该最佳映射法则对样本进行投影变换以达到类内间隔小类间间隔大的效果。相应地,为解决高维样本分类问题,进一步提出了该文算法的核化版本。在大量UCI数据集和Yale大学人脸数据库上的实验结果表明了该文算法的优越性。     

基于改进的监督LLE人脸识别算法

LLE是一种无监督的非线性降维方法,广泛应用于人脸特征提取,但是该方法缺乏样本点的类别信息.提出了一种新方法,在LLE的基础上引入有监督的学习机制和增加样本点的类别信息,通过减少类内距离而增加类间距离和最小化局部数据的全局重构误差,同时结合核邻域保持投影方法(KNPP)来提取高维人脸数据的非线性特征.算法有利于分类识别...     

应用监督近邻重构分析的高光谱遥感数据特征提取

针对高光谱遥感数据特征提取方法的研究,提出了一种新的监督近邻重构分析（Supervised Neighbor Reconstruction Analysis,SNRA）算法。该方法首先利用同一类别的近邻数据点对各数据点进行重构;然后在低维嵌入空间中保持该重构关系不变,尽可能地分离开非同类数据点,并利用总体散度矩阵来约束数据间的相关性;最后求解得到一个最佳投影矩阵,进而提取出鉴别特征。SNRA算法不仅保持了同类数据的局部结构而且增强了非同类数据的可分性,同时减少了数据的冗余信息。在Indian Pine和KSC高光谱遥感数据集上的实验结果表明:提出的方法能更好地揭示出高光谱遥感数据的内在特性,提取出更有效的鉴别特征,改善分类效果。     

Analytical form of globally optimal solution of weighted sum of intraclass separation and interclass separation

The Fisher linear discriminant analysis (FLDA)-based method is a common method for jointly optimizing the intraclass separation and the interclass separation of the projected feature vectors by defining the objective function as the ratio of the intraclass separation over the interclass separation. To address the eigenproblem of the FLDA, a quadratic equality constraint is imposed on the square of the \(l_{2}\) norm of the decision vector. However, the constrained optimization problem is highly nonconvex. This paper proposes to reformulate the objective function as a weighted sum of the intraclass separation and the interclass separation subject to the same quadratic equality constraint on the square of the \(l_{2}\) norm of the decision vector. Although both the objective function and the feasible set of the optimization problem are still nonconvex, this paper shows that the global minimum of the objective functional value is equal to the minimum singular value of the Hessian matrix of the objective function. Also, the globally optimal solution of the optimization problem is in the null space of the Hessian matrix minus this singular value multiplied by the identity matrix. As it is only required to find the singular value of the Hessian matrix, no numerical optimization-based computer aided design tool is required to find the globally optimal solution. Therefore, the globally optimal solution can be found in real time. Experimental results demonstrate the effectiveness of our proposed method.     

基于背景差分和混合帧差的运动目标检测

针对利用核密度估计建立背景模型时计算量大,运动目标和外界环境容易发生变化,提出一种基于改进的核密度估计背景差分法和改进的混合帧差法相结合的运动目标检测方法。该方法在背景建模时,先对背景差分后的图像进行分块和分类,并简化了核密度估计的核函数,对前景块中的像素进行核密度估计,减少了计算量。在混合帧差法中增加了动态阈值,提高了对光线变化的适应性。实验结果表明该方法能够完整地提取出运动目标,提高了目标检测的准确率。     

一种基于混合整型变换和3D-SPIHT的高光谱图像嵌入式无损压缩方法

利用高光谱图像空间、谱间相关性不同的特点,本文提出了一种基于混合整型变换和三维分层树集合划分算法(3D-SPIHT)的高光谱图像无损压缩方法,首先将波段进行分组,针对每一分组,采用不同的整型变换技术去相关,然后对变换系数进行系数重组,采用高效的基于小波系数特点的3D-SPIHT编码方法得到嵌入式码流,具有一定的抗误码性能.实验结果表明,该方法能够有效的去除高光谱图像的空间和谱间相关性,与现有的国际标准JPEG-LS、JPEG2000和基于三维离散余弦变换(3D-DCT)或三维离散小波变换(3D-DWT)的编码方法相比,压缩后的平均比特率均有明显降低.     

A comparative study of irregular pyramid matching in bag-of-bags of words model for image retrieval

In this paper, we assess three standard approaches to build irregular pyramid partitions for image retrieval in the bag-of-bags of words model that we recently proposed. These three approaches are: kernel \(k\)-means to optimize multilevel weighted graph cuts, normalized cuts and graph cuts, respectively. The bag-of-bags of words (BBoW) model is an approach based on irregular pyramid partitions over the image. An image is first represented as a connected graph of local features on a regular grid of pixels. Irregular partitions (subgraphs) of the image are further built by using graph partitioning methods. Each subgraph in the partition is then represented by its own signature. The BBoW model with the aid of graph extends the classical bag-of-words model, by embedding color homogeneity and limited spatial information through irregular partitions of an image. Compared with existing methods for image retrieval, such as spatial pyramid matching, the BBoW model does not assume that similar parts of a scene always appear at the same location in images of the same category. The extension of the proposed model to pyramid gives rise to a method we name irregular pyramid matching. The experiments on Caltech-101 benchmark demonstrate that applying kernel \(k\)-means to graph clustering process produces better retrieval results, as compared with other graph partitioning methods such as graph cuts and normalized cuts for BBoW. Moreover, this proposed method achieves comparable results and outperforms SPM in 19 object categories on the whole Caltech-101 dataset.     

Foreground object detection using top-down information based on em framework

In this paper, we present a novel foreground object detection scheme that integrates the top-down information based on the expectation maximization (EM) framework. In this generalized EM framework, the top-down information is incorporated in an object model. Based on the object model and the state of each target, a foreground model is constructed. This foreground model can augment the foreground detection for the camouflage problem. Thus, an object's state-specific Markov random field (MRF) model is constructed for detection based on the foreground model and the background model. This MRF model depends on the latent variables that describe each object's state. The maximization of the MRF model is the M-step in the EM framework. Besides fusing spatial information, this MRF model can also adjust the contribution of the top-down information for detection. To obtain detection result using this MRF model, sampling importance resampling is used to sample the latent variable and the EM framework refines the detection iteratively. Besides the proposed generalized EM framework, our method does not need any prior information of the moving object, because we use the detection result of moving object to incorporate the domain knowledge of the object shapes into the construction of top-down information. Moreover, in our method, a kernel density estimation (KDE)-Gaussian mixture model (GMM) hybrid model is proposed to construct the probability density function of background and moving object model. For the background model, it has some advantages over GMM- and KDE-based methods. Experimental results demonstrate the capability of our method, particularly in handling the camouflage problem.     

光谱角匹配加权核特征空间分离变换高光谱异常检测算法

提出了一种光谱角匹配(SAM)加权核特征空间分离变换(KEST)高光谱异常检测算法.在基于核的特征空间分离变换(KEST)算法基础上,利用光谱角匹配(SAM)测度对高维特征空间中检测点邻域差异相关矩阵(DCOR)中的每个样本引入权重因子,各样本权重因子取决于该样本光谱向量与检测窗口数据中心向量夹角,从而抑制检测窗口中的病态数据,突出主成分数据的贡献,使得DCOR矩阵能够更好地描述目标、背景数据分布差异.通过理论分析和对模拟、实际数据实验比较,证明该算法较传统异常检测算法和KEST算法具有更高的检测率.     

Ubiquitously Supervised Subspace Learning

In this paper, our contributions to the subspace learning problem are two-fold. We first justify that most popular subspace learning algorithms, unsupervised or supervised, can be unitedly explained as instances of a ubiquitously supervised prototype. They all essentially minimize the intraclass compactness and at the same time maximize the interclass separability, yet with specialized labeling approaches, such as ground truth, self-labeling, neighborhood propagation, and local subspace approximation. Then, enlightened by this ubiquitously supervised philosophy, we present two categories of novel algorithms for subspace learning, namely, misalignment-robust and semi-supervised subspace learning. The first category is tailored to computer vision applications for improving algorithmic robustness to image misalignments, including image translation, rotation and scaling. The second category naturally integrates the label information from both ground truth and other approaches for unsupervised algorithms. Extensive face recognition experiments on the CMU PIE and FRGC ver1.0 databases demonstrate that the misalignment-robust version algorithms consistently bring encouraging accuracy improvements over the counterparts without considering image misalignments, and also show the advantages of semi-supervised subspace learning over only supervised or unsupervised scheme.     

ITERATE: a conceptual clustering algorithm for data mining

The data exploration task can be divided into three interrelated subtasks: 1) feature selection, 2) discovery, and 3) interpretation. This paper describes an unsupervised discovery method with biases geared toward partitioning objects into clusters that improve interpretability. The algorithm ITERATE employs: 1) a data ordering scheme and 2) an iterative redistribution operator to produce maximally cohesive and distinct clusters. Cohesion or intraclass similarity is measured in terms of the match between individual objects and their assigned cluster prototype. Distinctness or interclass dissimilarity is measured by an average of the variance of the distribution match between clusters. The authors demonstrate that interpretability, from a problem-solving viewpoint, is addressed by the intraclass and interclass measures. Empirical results demonstrate the properties of the discovery algorithm and its applications to problem solving     

卷积神经网络（CNN）训练中卷积核初始化方法研究

本文提出了一种基于样本图像局部模式聚类的卷积核初始化方法,该方法可用于卷积神经网络（Convolutional neural network, CNN）训练中卷积核的初始化。在卷积神经网络中,卷积核的主要作用可看成是利用匹配滤波提取图像中的局部模式,并将其作为后续图像目标识别的特征。为此本文在图像训练集中选取一部分典型的样本图像,在这些图像中抽取与卷积核相同大小的子图作为图像局部模式矢量集合。首先对局部模式子图集合应用拓扑特性进行粗分类,然后对粗分类后的每一子类采用势函数聚类的方法获取样本图像中的典型局部模式子图,构成候选子图模式集,用它们作为CNN的初始卷积核进行训练。实验结果表明,本文方法可以明显加速CNN网络训练初期的收敛速度,同时对最终训练后的网络识别精度也有一定程度的提高。      

基于子块优化及全局整合的局部判别投影法

已有投影算法都直接通过完整的输入训练集求解最佳变换矩阵,难以进行增量式学习扩展。针对此问题,该文通过组合优化策略提出局部判别投影方法应用于分类问题。该算法同时包括类间判别信息和类内局部保持特征,求得的变换矩阵还具有正交性。此外,利用核函数将算法扩展至非线性应用,使之可以适应更多的数据类型。在ORL人脸库和小样本说话人辨认应用中验证了该算法的有效性。     

基于空间语义对象混合学习的复杂图像场景自动分类方法研究

场景分类是将多幅图像标记为不同语义类别的过程。该文针对现有方法对复杂图像场景分类性能欠佳的不足,提出一种新的基于空间语义对象混合学习的复杂图像场景分类方法。该方法以多尺度分割得到的图像对象而非整幅图像为主体进行产生式语义建模,统计各类有效特征挖掘对象的类别分布信息,并通过空间金字塔匹配,构建包含层次数据和语义信息的中间向量,弥补语义鸿沟的缺陷,训练中还结合判别式学习提高分类器的可信性。在实验数据集上的结果表明该方法具备较高的学习性能和分类精度,适用于多种类型和复杂内容图像的解译,具有较强的实用价值。     

Co-transduction for shape retrieval

In this paper, we propose a new shape/object retrieval algorithm, namely, co-transduction. The performance of a retrieval system is critically decided by the accuracy of adopted similarity measures (distances or metrics). In shape/object retrieval, ideally, intraclass objects should have smaller distances than interclass objects. However, it is a difficult task to design an ideal metric to account for the large intraclass variation. Different types of measures may focus on different aspects of the objects: for example, measures computed based on contours and skeletons are often complementary to each other. Our goal is to develop an algorithm to fuse different similarity measures for robust shape retrieval through a semisupervised learning framework. We name our method co-transduction, which is inspired by the co-training algorithm. Given two similarity measures and a query shape, the algorithm iteratively retrieves the most similar shapes using one measure and assigns them to a pool for the other measure to do a re-ranking, and vice versa. Using co-transduction, we achieved an improved result of 97.72% (bull's-eye measure) on the MPEG-7 data set over the state-of-the-art performance. We also present an algorithm called tri-transduction to fuse multiple-input similarities, and it achieved 99.06% on the MPEG-7 data set. Our algorithm is general, and it can be directly applied on input similarity measures/metrics; it is not limited to object shape retrieval and can be applied to other tasks for ranking/retrieval.     

改进Otsu法的目标图像分割

在现代战争中,快速而有效地对攻击目标进行识别和跟踪对获取战争主动权起着很重要的作用,而要达到这个目的,就需要从目标图像中准确地分割出目标.在分析Otsu的基础上,结合两类之间间距和类内距离对图像分割效果的影响,提出一种改进的最大类间方差法,并应用于目标图像的分割.实验结果分析显示该方法可达到较好的图像分割效果,较之Otsu方法更佳.     

基于双模态特征增强的目标检测算法研究与应用北大核心CSCD

为提升目标检测算法在复杂环境下的精确性和实用性,将多源信息和深度学习技术相结合,提出了一种基于双模态特征增强的目标检测方法。该方法以红外和可见光图像作为输入,利用颜色空间转换、边缘提取、直方图均衡化等传统图像处理方法丰富图像信息,达到数据增强效果;特征提取部分采用卷积神经网络结构分别提取目标红外及可见光信息,并设计混合注意力机制分别从通道和空间位置角度提升有效特征权重;同时,针对目标双模态信息,引入了自适应交叉融合结构,提高特征多样性;最后,利用交替上下采样将目标全局和局部特征充分融合,并以自主选择方式提取目标相关特征实现检测。通过在标准数据集以及实际场景数据集上的实验结果表明,所提方法有效融合并增强了目标多模态特征,提升了目标检测效果,并能较好的应用于电网场景中,辅助机器人完成目标设备检测。     

基于小波融合和PCA-核模糊聚类的遥感图像变化检测

本文提出了一种变化检测方法以提高算法的鲁棒性、检测精度以及抗噪性.首先对差值法构造的差异图和比值法构造的差异图进行小波融合.然后将融合图像分成互不重叠的小块,并用主成分分析得到图像块的正交基.通过将融合图像中每个像素的邻域小块映射到正交基上使得每个像素用一个特征向量来表示.最后用基于核的模糊C均值对特征向量进行聚类.实验结果显示与使用单一类型差异图的聚类方法相比,本方法由于采用了图像融合的策略而增强了鲁棒性,且由于采用了核模糊聚类,进一步提高了变化检测精度.此外由于使用了特征提取的技术,本方法具有一定的抗噪性能.     

基于时间一致性约束的核互相关器目标跟踪

针对相关滤波器跟踪算法在目标快速运动、遮挡和表观变化时易发生跟踪漂移或者丢失的问题,提出一种基于时间一致性和核互相关器的目标跟踪算法。该算法通过引入对图像噪声和杂波更具鲁棒性的核互相关向量,能够更精确地预测目标的仿射变化。同时,在学习过程中引入时间一致性约束,以解决因核相关器时间退化导致的跟踪漂移问题。最后,采用主灰度分量逆映射来提升跟踪器应对目标部分遮挡的能力。在公开的OTB100标准目标跟踪数据集中与提供的基准算法和其他性能更加先进的相关滤波算法进行对比,该算法平均跟踪速度为41f/s,相对fDSST和SAMF算法,其跟踪精度分别提升15.6%和6.4%,跟踪成功率分别提升33.3%和6.1%。实验结果表明,该算法在目标快速运动、遮挡或表观变化时仍能精确地跟踪目标。     

大规模MIMO相关信道下的联合天线分组和天线选择

在大规模多输入多输出系统中,针对密集部署的大型天线阵列之间的强相关性会抑制天线选择增益效果的问题。在系统下行链路场景下建立空间相关信道模型,提出了基于天线分组的天线选择算法。根据瞬时信道相关矩阵将天线阵列划分为若干组,保证各组内天线之间相关性较强。在完成天线分组的基础上,基于信道矩阵列范数准则在各组发射天线与接收天线之间构成的子信道矩阵中选择天线,进而构造有效发射天线与接收天线之间的信道矩阵。仿真分析了所提天线选择算法对系统遍历和速率的影响,结果表明,在基站天线数为32、接收天线数为2、选择天线数为2、天线相关因子为0.9的假设下,当信噪比为10 dB时,与基于相邻天线分组的天线选择算法相比,所提算法使系统和速率约提高了27.5%,且所提算法若要与最优天线选择算法达到相同的和速率,仅需将其信噪比提升1～2 dB即可。