基于改进的局部保持投影高光谱图像分类研究*

针对高光谱图像分类中基于流形的降维方法进行了研究。提出一种改进的局部保持投影（LPP）方法即MLPP方法。该方法利用标签信息避免了传统LPP在邻接图构建中很难确定的邻域大小的选择问题,同时采用更能反映高维数据间相关性的统计特征量相关系数来衡量数据之间的相似程度。设计的权重矩阵既保持类内数据的几何结构,又最大化类间距离。而且MLPP不依赖任何参数和先验知识。在两个高光谱图像上的实验结果表明MLPP增加了不同光谱特征地物之间的可分性,在提高分类性能上明显优于其他传统的降维方法。     

An Algorithm for Enumerating All Spanning Trees of a Directed Graph

We present an O(NV + V ³ ) time algorithm for enumerating all spanning trees of a directed graph. This improves the previous best known bound of O(NE + V+E) [1] when V ² =o(N) , which will be true for most graphs. Here, N refers to the number of spanning trees of a graph having V vertices and E edges. The algorithm is based on the technique of obtaining one spanning tree from another by a series of edge swaps. This result complements the result in the companion paper [3] which enumerates all spanning trees in an undirected graph in O(N+V+E) time. Received September 11, 1997; revised March 6, 1998.     

A Graph‐based Approach to Continuous Line Illustrations with Variable Levels of Detail

This paper introduces a method for automatically generating continuous line illustrations, drawings consisting of a single line, from a given input image. Our approach begins by inferring a graph from a set of edges extracted from the image in question and obtaining a path that traverses through all edges of the said graph. The resulting path is then subjected to a series of post‐processing operations to transform it into a continuous line drawing. Moreover, our approach allows us to manipulate the amount of detail portrayed in our line illustrations, which is particularly useful for simplifying the overall illustration while still retaining its most significant features. We also present several experimental results to demonstrate that our approach can automatically synthesize continuous line illustrations comparable to those of some contemporary artists.     

On simultaneous straight-line grid embedding of a planar graph and its dual

Simultaneous representations of planar graphs and their duals normally require that the dual vertices to be placed inside their corresponding primal faces, and the edges of the dual graph to cross only their corresponding primal edges. Erten and Kobourov [C. Erten, S.G. Kobourov, Simultaneous embedding of a planar graph and its dual on the grid, Theory Computer Systems 38 (2005) 313-327] provided a linear time algorithm on simultaneous straight-line grid embedding of a 3-connected planar graph and its dual such that all the vertices are placed on grid points and each edge is drawn as one straight-line segment except for one which is drawn using two segments. Their drawing size is (2n−2)×(2n−2), where n is the total number of vertices in the graph and its dual. They raised an open question on whether there is a large class of planar graphs that allows this simultaneous straight-line grid embedding on a smaller grid. We answer this open question by giving a linear time simultaneous straight-line grid embedding algorithm for a 3-connected planar graph and its dual on a grid of size (n−1)×n.     

一种基于多视立体视觉的多视图直线匹配方法

提出了一种基于多视立体视觉（Multiple view stereo,MVS）进行多视图直线匹配的方法. 本文方法首先利用MVS所得到的三维点云及其可见性信息,建立三维点与图像直线的对应关系. 根据此对应关系,为每条图像直线建立由一个三维点集和一个三维单位向量构成的描述子,用以衡量图像直线之间的相似性及一致性. 之后,本文方法以所有图像直线为顶点建立一个图,并引入了图谱分析来获取统一的顶点距离度量. 最后,本方法对DBSCAN聚类算法进行了修改,并用修改后的算法从图谱分析结果中获取可靠的直线匹配. 实验显示,本方法比已有方法更加鲁棒,并且有更高的准确率.     

基于双属性图表示的通用人脸图像识别系统

该文提出一个通用的人脸图像识别系统GFRS,对于人脸库中每个待识目标只存储一幅图像的情况下,可以实现各种干扰条件下人脸图像的识别,该系统提出使用双属性图来表示人脸图像,首先利用逐步求精定位法得到人脸图像各局部特征点的位置,图中每个特征点由两上特征属性来描述,即局部主成分特征系数和Gabor变换特征参数,从而构造了双属性图,然后给出双属性图匹配函数、匹配算法以及相应的识别方法。     

基于线图与PSO的网络重叠社区发现

从优化模块度的角度出发,引入线图理论,给出线图的硬划分与原 图的有重叠划分相对应的理论证明, 提出了一种基于线图与粒子群优化技术的网络重叠社区发现算法(Communities discovery based on line graph and particle swarm optimization, LGPSO), 该方法通过粒子群优化 (Particle swarm optimization, PSO)算法寻找网络对应线图的最优划分来发现网络重叠社区, 实验结果显示,该方法能够在无先验信息的条件下快速有效地揭示网络的重叠社区结构.     

Finding the k Shortest Paths in Parallel

A concurrent-read exclusive-write PRAM algorithm is developed to find the k shortest paths between pairs of vertices in an edge-weighted directed graph. Repetitions of vertices along the paths are allowed. The algorithm computes an implicit representation of the k shortest paths to a given destination vertex from every vertex of a graph with n vertices and m edges, using O(m+nk log ² k) work and O( log^3k log ^*k+ log n( log log k+ log ^*n)) time, assuming that a shortest path tree rooted at the destination is pre-computed. The paths themselves can be extracted from the implicit representation in O( log k + log n) time, and O(n log n +L) work, where L is the total length of the output. Received July 2, 1997; revised June 18, 1998.     

确定平面直线图完全单调链集的算法

本文提出求平面直线图完全单调链集的一种算法。基本思想是,先求平面直线图Ｇ的顶点集的凸壳及其直径,然后求各顶点在直径上垂直投影点,并按投影点的ｘ（或ｙ）坐 标排序Ｇ的顶点,最后按一定规则找出完全单调链集。     

基于三角剖分对偶图的结构表示模型及其算法

通过提取中轴骨架并对其进行跟踪而获得层次结构图是模式识别中最常用的一种方法。论文提出一种相对骨架的新型结构表示模型“三角剖分对偶图”。利用这种结构表示模型可以从二值图像中直接获得层次结构图,并表达为树的最基本的数据结构。此外,论文还给出了三角剖分对偶图的提取和优化算法,并与骨架表示模型的层次结构图作了比较。     

Super-connected and hyper-connected jump graphs

Let G be a connected graph. The graph G is said to be super-connected if for every minimum vertex cut S of G, G?S has isolated vertices. Moreover, it is said to be hyper-connected if for every minimum vertex cut S, G?S has exactly two components, one of which is an isolated vertex. In this note, we give a necessary and sufficient condition for a graph G whose jump graph J(G) (the complement of line graph of G) is, respectively, super-connected and hyper-connected.     

Algorithms for a Class of Isotonic Regression Problems

The isotonic regression problem has applications in statistics, operations research, and image processing. In this paper a general framework for the isotonic regression algorithm is proposed. Under this framework, we discuss the isotonic regression problem in the case where the directed graph specifying the order restriction is a directed tree with n vertices. A new algorithm is presented for this case, which can be regarded as a generalization of the PAV algorithm of Ayer et al. Using a simple tree structure such as the binomial heap, the algorithm can be implemented in O(n log n) time, improving the previously best known O(n ² ) time algorithm. We also present linear time algorithms for special cases where the directed graph is a path or a star. Received September 2, 1997; revised January 2, 1998, and February 16, 1998.     

Efficient Orthogonal Drawings of High Degree Graphs

Most of the work that appears in the two-dimensional orthogonal graph drawing literature deals with graphs whose maximum degree is four. In this paper we present an algorithm for orthogonal drawings of simple graphs with degree higher than four. Vertices are represented by rectangular boxes of perimeter less than twice the degree of the vertex. Our algorithm is based on creating groups / pairs of vertices of the graph. The orthogonal drawings produced by our algorithm have area at most (m-1) ( m / 2 +2) . Two important properties of our algorithm are that the drawings exhibit a small total number of bends (less than m ), and that there is at most one bend per edge. Received January 15, 1997; revised February 1, 1998.     

Knowledge representation for mathematical discovery: Three experiments in graph theory

This paper describes the nature of mathematical discovery (including concept definition and exploration, example generation, and theorem conjecture and proof), and considers how such an intelligent process can be simulated by a machine. Although the material is drawn primarily from graph theory, the results are immediately relevant to research in mathematical discovery and learning.The thought experiment, a protocol paradigm for the empirical study of mathematical discovery, highlights behavioral objectives for machine simulation. This thought experiment provides an insightful account of the discovery process, motivates a framework for describing mathematical knowledge in terms of object classes, and is a rich source of advice on the design of a system to perform discovery in graph theory. The evaluation criteria for a discovery system, it is argued, must include both a set of behavior to display (behavioral objectives) and a target set of facts to be discovered (factual knowledge).Cues from the thought experiment are used to formulate two hierarchies of representational languages for graphy theory. The first hierarchy is based on the superficial terminology and knowledge of the thought experiment. Generated by formal grammars with set-theoretic semantics, this eminently reasonable approach ultimately fails to meet the factual knowledge criteria. The second hierarchy uses declarative expressions, each of which has a semantic interpretation as a stylized, recursive algorithm that defines a class by generating it correctly and completely. A simple version of one such representation is validated by a successful, implemented system called Graph Theorist (GT) for mathematical research in graph theory. GT generates correct examples, defines and explores new graph theory properties, and conjectures and proves theorems.Several themes run through this paper. The first is the dual goals, behavioral objectives and factural knowledge to be discovered, and the multiplicity of their demands on a discovery system. The second theme is the central role of object classes to knowledge representation. The third is the increased power and flexibility of a constructive (generator) definition over the more traditional predicate (tester) definition. The final theme is the importance of examples and recursion in mathematical knowledge. The results provide important guidance for further research in the simulation of mathematical discovery.     

描述工程图纸扫描图象的单义域邻接图的构建方法

工程图纸扫描图象的编码是矢量化中一个重要问题。该文首先对二值图象作水平黑游程编码,建立游程部接图来表达图象,然后采用深度优先遍历邻接图,即相关游程基于宽度与拓扑的一致性组成条形域（图的一个路径）,分裂其中的多义域得到单义域（线段域和圆弧域）。图象的描述单元变为单义域,再继承游程的拓扑关系,建立单义域邻接图。单义域邻接图较好地描述了图象几何属性和拓扑关系。最后给出基于单义域邻接图的矢量化应用,效果也较好。     

宽带图像信号的直线提取

该文用FPGA＋DSP实现了宽带图像信号的实时直线提取,适用于通常的视频图像信号和线阵CCD成像的连续图像信号。实现的直线提取模块已经用于高传输率的快速遥感图像处理系统。算法将直线提取中的参数计算分解为一个可以累加的过程,从而简化了计算,同时保证了直线参数的准确性,在理论上证明了算法的完备性,文章的最后给出了实验结果及比较。     

A hierarchical recurrent approach to predict scene graphs from a visual‐attention‐oriented perspective

A scene graph provides a powerful intermediate knowledge structure for various visual tasks, including semantic image retrieval, image captioning, and visual question answering. In this paper, the task of predicting a scene graph for an image is formulated as two connected problems, ie, recognizing the relationship triplets, structured as <subject‐predicate‐object>, and constructing the scene graph from the recognized relationship triplets. For relationship triplet recognition, we develop a novel hierarchical recurrent neural network with visual attention mechanism. This model is composed of two attention‐based recurrent neural networks in a hierarchical organization. The first network generates a topic vector for each relationship triplet, whereas the second network predicts each word in that relationship triplet given the topic vector. This approach successfully captures the compositional structure and contextual dependency of an image and the relationship triplets describing its scene. For scene graph construction, an entity localization approach to determine the graph structure is presented with the assistance of available attention information. Then, the procedures for automatically converting the generated relationship triplets into a scene graph are clarified through an algorithm. Extensive experimental results on two widely used data sets verify the feasibility of the proposed approach.     

On equitable colouring of Knödel graphs

The notion of equitable colouring was introduced by Meyer in 1973. In this paper we find the equitable chromatic number χ₌ for the line graph of Knödel graphs L(W_{Δ, n}), central graph of Knödel graphs C(W_{Δ, n}) and corona graph of Knödel graphs W_{Δ, m}° W_{Δ, n}. As a by-product we obtain a new class of graphs that confirm Equitable Colouring Conjecture.     

基于图形箭头符号识别的直线图形自动生成方法

提出了一种基于图形箭头符号识别的直线图形自动生成方法。在箭头识别的基础上 ,进行尺寸信息的提取 ,再以“尺寸标注线的箭头两端为中心的小区域内有尺寸界限线”为前提 ,根据尺寸信息 ,通过对尺寸界限线的识别、理解 ,形成对图形的约束及组织 ,从而实现标准图形的自动输出。     

基于点线结合特征的单目视觉里程计

SLAM(即时定位与地图构建)系统是近年来计算机视觉领域的一大重要课题,其中特征法的SLAM凭借稳定性好、计算效率高的优点成为SLAM算法的主流。目前特征法SLAM主要基于点特征进行。针对基于点特征的视觉里程计依赖于数据质量,相机运动过快时容易跟丢,且生成的特征地图不包含场景结构信息等缺点,提出了一种基于点线结合特征的优化算法。相较于传统基于线段端点的六参数表达方式,算法采用一种四参数的方式表示空间直线,并使用点线特征进行联合图优化估计相机位姿。使用公开数据集和自采集鱼眼影像数据分别进行实验的结果表明,与仅使用点特征的方法相比,该方法可有效改善因相机运动过快产生的跟丢问题,增加轨迹长度,提升位姿估计精度,且生成的稀疏特征地图更能反映场景结构特征。