平面点集Delaunay三角剖分的分治算法

为发展图形网格化技术,研究了平面点集的三角剖分算法.根据经典算法中在实际应用中遇到的共性问题,提炼了3个工具算法;为了更好地表示平面区域划分的拓扑信息,引入了双链接边表(DCEL)的数据结构.在此基础上,设计并实现了平面集Delaunay三角剖分分治算法,并对特殊退化情况进行了处理,通过计算表明了该算法时间复杂度为0(N* logN).实验数据结果验证了该算法的正确性、健壮性.     

A moving object segmentation algorithm for static camera via active contours and GMM

Moving object segmentation is one of the most challenging issues in computer vision. In this paper, we propose a new algorithm for static camera foreground segmentation. It combines Gaussian mixture model (GMM) and active contours method, and produces much better results than conventional background subtraction methods. It formulates foreground segmentation as an energy minimization problem and minimizes the energy function using curve evolution method. Our algorithm integrates the GMM background model, shadow elimination term and curve evolution edge stopping term into energy function. It achieves more accurate segmentation than existing methods of the same type. Promising results on real images demonstrate the potential of the presented method. Supported by National Basic Research Program of China (Grant No. 2006CB303105), the Chinese Ministry of Education Innovation Team Fund Project (Grant No. IRT0707), the National Natural Science Foundation of China (Grant Nos. 60673109 and 60801053), Beijing Excellent Doctoral Thesis Program (Grant No. YB20081000401), Beijing Municipal Natural Science Foundation (Grant No. 4082025), and Doctoral Foundation of China (Grant No. 20070004037)     

A dynamic integration algorithm to model surfaces from multiple range views

This paper presents a dynamic integration algorithm to triangulate a surface from multiple range views. This integration technique is based on the reparameterization of the canonic subsets of the Venn diagram of the set of range views. We compute a model based onN views from a model based onN–1 views by processing only the surface segments visible in theN ^th view. An experimental result shows that the proposed integration algorithm can process complex multipart objects containing holes.     

A counterexample to an algorithm for computing monotone hulls of simple polygons

A two-stage algorithm was recently proposed by Sklansky (1982) for computing the convex hull of a simple polygon P. The first step is intended to compute a simple polygon $\text{P}{}^1$ which is monotonic in both the x and y directions and which contains the convex hull vertices of P. The second step applies a very simple convex hull algorithm on $\text{P}{}^1$. In this note we show that the first step does not always work correctly and can even yield non-simple polygons, invalidating the use of the second step. It is also shown that the first step can discard convex hull vertices thus invalidating the use of any convex hull algorithm in the second step.     

A parallel algorithm for surface-based object reconstruction

This paper presents a parallel algorithm that approximates the surface of an object from a collection of its planar contours. Such a reconstruction has wide applications in such diverse fields as biological research, medical diagnosis and therapy, architecture, automobile and ship design, and solid modeling. The surface reconstruction problem is transformed into the problem of finding a minimum-cost acceptable path on a toroidal grid graph, where each horizontal and each vertical edge have the same orientation. An acceptable path is closed path that makes a complete horizontal and vertical circuit. We exploit the structure of this graph to develop efficient parallel algorithms for a message-passing computer. Givenp processors and anm byn toroidal graph, our algorithm will find the minimum cost acceptable path inO(mn log(m)/p) steps, ifp =O(mn/((m + n) log(mn/(m + n)))), which is an optimal speedup. We also show that the algorithm will sendO(p ²(m + n)) messages. The algorithm has a linear topology, so it is easy to embed the algorithm in common multiprocessor architectures.     

基于动态奖惩的CDCL SAT求解器分支启发式算法

分支启发式算法在CDCL SAT求解器中有着非常重要的作用,传统的分支启发式算法在计算变量活性得分时只考虑了冲突次数而并未考虑决策层和冲突决策层所带来的影响。为了提高SAT问题的求解效率,受EVSIDS和ACIDS的启发,提出了基于动态奖惩DRPB的分支启发式算法。每当冲突发生时,DRPB通过综合考虑冲突次数、决策层、冲突决策层和变量冲突频率来更新变量活性得分。用DRPB替代VSIDS算法改进了Glucose 3.0,并测试了SATLIB基准库、2015年和2016年SAT竞赛中的实例。实验结果表明,与传统、单一的奖励变量分支策略相比,所提分支策略可以通过减少搜索树的分支和布尔约束传播次数来减小搜索树的规模并提高SAT求解器的性能。     

A variable neighborhood search heuristic algorithm for the double vehicle routing problem with multiple stacks

This paper addresses the double vehicle routing problem with multiple stacks (DVRPMS) in which a fleet of vehicles must collect items in a pickup region and then travel to a delivery region where all items are delivered. The load compartment of all vehicles is divided into rows (horizontal stacks) of fixed profundity (horizontal heights), and on each row, the unloading process must respect the last‐in‐first‐out policy. The objective of the DVRPMS is to find optimal routes visiting all pickup and delivery points while ensuring the feasibility of the vehicle loading plans. We propose a new integer linear programming formulation, which was useful to find inconsistencies in the results of exact algorithms proposed in the literature, and a variable neighborhood search based algorithm that was able to find solutions with same or higher quality in shorter computational time for most instances when compared to the methods already present in the literature.     

A type-merging algorithm for extracting an isosurface from volumetric data

A new approach for reducing the number of triangles representing an isosurface in volumetric data is presented. The basic idea is to classify the configurations of the marching cubes approach into types. Surface patches traversing neighboring cubes of the same type can be merged into patches, which can be approximated with fewer and larger triangles. Experimental results show that the number of triangles is about 50% of that obtained with the marching cubes algorithm, with comparable image quality. The execution time is somewhat longer than that of the marching cubes algorithm.     

Mesh filtering algorithm using an adaptive 3D convolution kernel applied to a volume-based vector distance field

This paper addresses the problem of feature preserving mesh filtering, which occurs in surface reconstruction of scanned objects, which include acquisition noise to be removed without altering sharp edges. We propose a method based on a vector field distance transform of the mesh to process. It is a volume-based implicit surface modeling, which provides an alternative representation of meshes. We use an adaptive 3D convolution kernel applied to the voxels of the distance transform model. Weights of the kernel elements are determined according to the angle between the vectors of the implicit field. We also propose a new adaptation of the Marching Cubes algorithm in order to extract the isosurface from the vector implicit field after the filtering process. We compare our method to the previous one introduced using the vector field representation and to other feature preserving adaptive filtering algorithms. According to error metric evaluations, we show that our new design provides high quality filtering results while better preserving geometric features.     

A classification method using a hybrid genetic algorithm combined with an adaptive procedure for the pool of ellipsoids

This paper presents a hybrid classification method that utilizes genetic algorithms (GAs) and adaptive operations of ellipsoidal regions for multidimensional pattern classification problems with continuous features. The classification method fits a finite number of the ellipsoidal regions to data pattern by using hybrid GAs, the combination of local improvement procedures and GAs. The local improvement method adaptively expands, rotates, shrinks, and/or moves the ellipsoids while each ellipsoid is separately handled with a fitness value assigned during the GA operations. A set of significant features for the ellipsoids are automatically determined in the hybrid GA procedure by introducing “don’t care” bits to encode the chromosomes. The performance of the method is evaluated on well-known data sets and a real field classification problem originated from a deflection yoke production line. The evaluation results show that the proposed method can exert superior performance to other classification methods such as k nearest neighbor, decision trees, or neural networks. Ki K. Lee received the B.S. degree from Han Yang University, Seoul, Korea in 1994, and the M.S. and Ph.D. degrees in industrial engineering from Korea Advanced Institute Science and Technology (KAIST), Daejeon, Korea in 1996 and 2005, respectively. From 2001 to 2004, he was a senior research engineer in telecommunication systems laboratory of LG Electronics Inc. Since 2005, he has been an assistant professor in the School of Management at Inje University, Kimhae, Korea. His research interests include intelligent decision support systems, soft computing, and pattern recognition. Wan C. Yoon received the B.S. degree from Seoul National University, Korea in 1977, the M.S. degree from KAIST, Korea in 1979, and the Ph.D. degree in industrial and systems engineering from Georgia Institute of Technology in 1987. He is professor of the Department of Industrial Engineering at KAIST, Korea. His research interests include application of artificial intelligence, human decision-making and aiding, information systems, and joint intelligent systems. Dong H. Baek received the B.S. degree from Han Yang University, Seoul, Korea in 1992, and the M.S. and Ph.D. degrees in industrial engineering from Korea Advanced Institute Science and Technology (KAIST), Daejeon, Korea in 1994 and 1999, respectively. He is an assistant professor in management information systems at department of business administration, Hanyang University, Korea. His research interests include management information systems, system engineering, and machine learning.     

A Method for Constructing Interpolatory Subdivision Schemes and Blending Subdivisions

This paper presents a universal method for constructing interpolatory subdivision schemes from known approximatory subdivisions. The method establishes geometric rules of the associated interpolatory subdivision through addition of further weighted averaging operations to the approximatory subdivision. The paper thus provides a novel approach for designing new interpolatory subdivision schemes. In addition, a family of subdivision surfaces varying from the given approximatory scheme to its associated interpolatory scheme, namely the blending subdivisions, can also be established. Based on the proposed method, variants of several known interpolatory subdivision schemes are constructed. A new interpolatory subdivision scheme is also developed using the same technique. Brief analysis of a family of blending subdivisions associated with the Loop subdivision scheme demonstrates that this particular family of subdivisions are globally C¹ continuous while maintaining bounded curvature for regular meshes. As a further extension of the blending subdivisions, a volume‐preserving subdivision strategy is also proposed in the paper.