Computing point-set surfaces with controlled spatial variation of residuals

This paper presents an accurate method for computing point-set surfaces from input data that can suppress the noise effect in the resulting point-set surface. This is accomplished by controlling spatial variation of residual errors between the input data and the resulting point-set surface and offsetting any systematic bias. More specifically, this method first reduces random noise of input data based on spatial autocorrelation statistics: the statistics Z via Moran’s I. The bandwidth of the surface is adjusted until the surface reaches desired value of the statistics Z corresponding to a given significance level. The method then compensates for potential systematic bias of the resultant surface by offsetting along computed normal vectors. Computational experiments on various point sets demonstrate that the method leads to an accurate surface with controlled spatial variation of residuals and reduced systematic bias.     

Automated reasoning about elementary point-set topology

In this paper we present first-order formulas for basic point-set topology, in an attempt to extend the mathematical range available for exploration with automated theorem-proving programs. We present topology definitions and sample lemmas both in first-order logic and in clausal form. We then illustrate some of the difficulties of these sample lemmas through a proof of a basic lemma in five parts.This work was supported by the Applied Mathematical Sciences subprogram of the Office of Energy Research, U.S. Department of Energy, under contract W-31-109-Eng-38, and by the Division of Educational Programs, Argonne National Laboratory.     

Optimal parallel algorithms for point-set and polygon problems

In this paper we give parallel algorithms for a number of problems defined on point sets and polygons. All our algorithms have optimalT(n) * P(n) products, whereT(n) is the time complexity andP(n) is the number of processors used, and are for the EREW PRAM or CREW PRAM models. Our algorithms provide parallel analogues to well-known phenomena from sequential computational geometry, such as the fact that problems for polygons can oftentimes be solved more efficiently than point-set problems, and that nearest-neighbor problems can be solved without explicitly constructing a Voronoi diagram.The research of R. Cole was supported in part by NSF Grants CCR-8702271, CCR-8902221, and CCR-8906949, by ONR Grant N00014-85-K-0046, and by a John Simon Guggenheim Memorial Foundation fellowship. M. T. Goodrich's research was supported by the National Science Foundation under Grant CCR-8810568 and by the National Science Foundation and DARPA under Grant CCR-8908092.     

Optimal parallel algorithms for point-set and polygon problems

In this paper we give parallel algorithms for a number of problems defined on point sets and polygons. All our algorithms have optimalT(n) * P(n) products, whereT(n) is the time complexity andP(n) is the number of processors used, and are for the EREW PRAM or CREW PRAM models. Our algorithms provide parallel analogues to well-known phenomena from sequential computational geometry, such as the fact that problems for polygons can oftentimes be solved more efficiently than point-set problems, and that nearest-neighbor problems can be solved without explicitly constructing a Voronoi diagram.     

Algorithms for bichromatic line-segment problems and polyhedral terrains

We consider a variety of problems on the interaction between two sets of line segments in two and three dimensions. These problems range from counting the number of intersecting pairs between m blue segments andn red segments in the plane (assuming that two line segments are disjoint if they have the same color) to finding the smallest vertical distance between two nonintersecting polyhedral terrains in three-dimensional space. We solve these problems efficiently by using a variant of the segment tree. For the three-dimensional problems we also apply a variety of recent combinatorial and algorithmic techniques involving arrangements of lines in three-dimensional space, as developed in a companion paper.Work on this paper by the first author has been supported in part by the National Science Foundation under Grant CCR-9002352. Work by the second author was supported in part by the National Science Foundation under Grant CCR-8714565. The fourth author has been supported in part by the Office of Naval Research under Grant N0014-87-K-0129, by the National Science Foundation under Grant NSF-DCR-83-20085, by grants from the Digital Equipment Corporation and the IBM Corporation, and by a grant from the US-Israeli Binational Science Foundation.     

View matching with blob features

This article introduces a new region based feature for object recognition and image matching. In contrast to many other region based features, this one makes use of colour in the feature extraction stage. We perform experiments on the repeatability rate of the features across scale and inclination angle changes, and show that avoiding to merge regions connected by only a few pixels improves the repeatability. We introduce two voting schemes that allow us to find correspondences automatically, and compare them with respect to the number of valid correspondences they give, and their inlier ratios. We also demonstrate how the matching procedure can be applied to colour correction.     

Hierarchical generalized Hough transforms and line-segment based generalized Hough transforms

Generalized Hough transforms are fast point pattern matching algorithms which have applications in image processing. Two extensions to generalized Hough transforms are described. The first allows for matching hierarchically organized point patterns. An example concerning image registration using hierarchical point patterns is described. The second extension concerns matching patterns of simple geometric objects, such as line segments, rather than patterns of points.     

A new graph matching method for point-set correspondence using the EM algorithm and Softassign

Finding correspondences between two point-sets is a common step in many vision applications (e.g., image matching or shape retrieval). We present a graph matching method to solve the point-set correspondence problem, which is posed as one of mixture modelling. Our mixture model encompasses a model of structural coherence and a model of affine-invariant geometrical errors. Instead of absolute positions, the geometrical positions are represented as relative positions of the points with respect to each other. We derive the Expectation–Maximization algorithm for our mixture model. In this way, the graph matching problem is approximated, in a principled way, as a succession of assignment problems which are solved using Softassign. Unlike other approaches, we use a true continuous underlying correspondence variable. We develop effective mechanisms to detect outliers. This is a useful technique for improving results in the presence of clutter. We evaluate the ability of our method to locate proper matches as well as to recognize object categories in a series of registration and recognition experiments. Our method compares favourably to other graph matching methods as well as to point-set registration methods and outlier rejectors.     

Dynamical features of GPS PWV variation associated with lightning activity

Water vapour is one component that causes spatial and temporal fluctuations in the lower atmosphere, which, in turn, introduce time delays into the global positioning system (GPS) signal. These delays can be exploited to estimate the precipitable water vapour (PWV), which is beneficial for meteorological applications. Because the vertical transfer of warm air to the troposphere triggers instability and atmospheric charges, lightning phenomena can fundamentally affect the GPS signals through changes in water vapour. From this hypothesis, GPS-derived PWV data have been proposed as a tool for monitoring lightning activity. In this paper, the variation of PWV in days with lightning at four selected stations in Peninsular Malaysia during the past two inter-monsoon events in May and November 2009 was observed. To verify the response, a peak alignment method was proposed with regard to the lightning with more than 100 discharge events and lightning days from 07:00 to 20:00 LT. A total of 66 lightning series were assessed, and PWV was observed to increase by approximately 2.46 mm and decrease by 4.16 mm before and after the peak point, respectively, when compared to fair-weather conditions. Approximately 62% of the lightning start times were concentrated within a period of less than 1 h before or after the PWV reached a maximum peak, 24% were observed between 1–2 h, and 14% started after 2 h. This trend implies that the use of GPS PWV data was more consistent and robust for the detection of lightning activity compared to the use of surface temperature and relative humidity data.     

End-to-end music emotion variation detection using iteratively reconstructed deep features

Multimedia Tools and Applications - Automatic music emotion recognition (MER) has received increased attention in areas of music information retrieval and user interface development. Music emotion...     

View independent face detection based on horizontal rectangular features and accuracy improvement using combination kernel of various sizes

This paper proposes a view independent face detection method based on horizontal rectangular features, and accuracy improvement by combining kernels of various sizes. Since the view changes of faces induce large variation in appearance in the horizontal direction, local kernels are applied to horizontal rectangular regions to model such appearance changes. Local kernels are integrated by summation, and then used as a summation kernel for support vector machine (SVM). View independence is shown to be realized by the integration of local horizontal rectangular kernels. However, in general, local kernels (features) of various sizes have different similarity measures, such as detailed and rough similarity, and thus their error patterns are different. If the local and global kernels are combined well, the generalization ability is improved. This research demonstrates the comparative effectiveness of combining the global kernel and local kernels of various sizes as a summation kernel for SVM against use of only the global kernel, only the combination of local kernels and Adaboost with SVMs with a kind of local kernel.     

专家视图与本体视图的语义映射方法

由于本体具有强大的知识表示和推理能力,本体已经在很多领域得到了广泛应用.然而,本体的深入应用还面临着很多深层次的共性语义映问题,已有的本体建模方法仅仅提出了一些简单的指导原则和基本步骤,使得知识工程师仍然无从下手.在基于领域专家知识构建领域语义知识库时,针对领域专家知识中的一词多义、多元关系和安全需求等3类共性语义映射问题,从5个方面展开了深入研究,提出了相应的语义映射方法,总结了10条本体建模约定.最后构建了一个完整的应用案例,并对所提出的5类语义映射方法进行了评价.     

视图Morphing技术在视图插补中的应用

图像Morphing是图像之间插补变换的一类被广泛使用的技术，这种方法是基于两幅图像之间的像素位置和颜色的内插，但对3D场景或物体，它不能保证所生成图像的直实性，即不满足透视性质，在此介绍一种图像插补变换的新方法：视图Morphing技术，它可以保证生成图像的真实性，且不需要知道图像的3D信息，这避免了图像的3D重建，因而具有极大的应用价值。一般的视图Morphing技术需要知道图像的投影矩阵，可是用目前的技术，从图像直接获得投影矩阵是很困难的。视图Morphing技术事先不需要知道图像的投影矩阵，就可以对摄像机刚体运动所获得的两幅图像进行插补运算，获得中间视点的图像。     

粗糙集信息观与代数观联系研究

粗糙集的信息观和代数观是通过不同的方式来定义知识的,两种观点在绝对约简中完全等价,但是在相对约简中,两种观点的关系并不明确。文章研究了两种观点在相对约简中的关系,证明了两种观点在相容决策表中完全相同,而在不相容决策表中,信息观的现有定理存在错误,也就是说信息观不适用于不相容决策表。     

视图Morphing技术在视图插补中的应用

图像 Morphing是图像之间插补变换的一类被广泛使用的技术,这种方法是基于两幅图像之间的像素位置和颜色的内插,但对3D场景或物体,它不能保证所生成图像的真实性,即不满足透视性质,在此介绍一种图像插补变换的新方法:视图Morphing 技术,它可以保证生成图像的真实性,且不需要知道图像的3D信息,这避免了图像的3D重建,因而具有极大的应用价值.一般的视图 Morphing技术需要知道图像的投影矩阵,可是用目前的技术,从图像直接获得投影矩阵是很困难的.视图Morphing技术事先不需要知道图像的投影矩阵,就可以对摄像机刚体运动所获得的两幅图像进行插补运算,获得中间视点的图像.     

对象视图与外模式

视图和外模式都是数据库技术中很有用的基本概念，它们的共同特征是可用来对数据库模式进行抽象。本文讨论了在面向对象模型中引入视图机制的一些问题，并提出了用视图构成数据库外模式的初步方案。     

Cognitive View of Reuse and Redesign

Reusable components are not enough. Program designers need tools that help them understand the components and how to use them. Fortunately, some support tools do exist.