一种新的加权最小二乘测距定位方法

三站二维定位问题是最小定位问题，一般只利用几何关系就能获得定位估计，但由于未利用距离测量的统计信息，定位精度较差。该文先将点到点的距离测量转化为点到线的距离估计，再在此基础上推导出一种新的加权几休定位方法，其中加权的确定利用了距离测量的统计信息。仿真实验表明，在测距误差较小时新方法具有更高的定位精度。     

Detecting the wheel pattern of a vehicle using stereo images

This paper presents a method for detecting the wheels of a vehicle in stereo image pairs. The method consists of two steps: (i) geometrical transformation; and (ii) circle extraction. The geometrical transformation uses the disparity values obtained from a stereo image pair to calculate the parameters of the plane containing wheels of the vehicle. By using these parameters, we transform any elliptical wheels contained in the plane to circular ones which can be extracted by the circle extraction algorithm. The circle extraction algorithm consists of (1) template matching and (2) Hough transform. In order to save computation and improve the results in the Hough transform, we employ two constraints (a) the neighbor-region edge connectivity and (b) the gradient direction of each edge point, to eliminate non-circular edge points. Experimental results show that these two constraints do eliminate non-circular edge points and preserve any circle embedded in edges. From the final results, we can observe that our method can detect and locate the wheels of a vehicle successfully.     

M-P神经元模型的几何意义及其应用

给出M-P神经元模型的几何意义,这个几何的铨释,给神经元一个非常直观的理解,利用这个直观的理解,给出两个颇为有趣的应用:(1)用此法给出三层前向神经网络的学习能力的基本定理的新的证明;(2)给出前向网络的拓扑结构设计的新方法.     

Freeform texture representation and characterisation based on triangular mesh projection techniques

Texture characterisation for freeform non-Euclidean surfaces is becoming increasingly important due to the widespread of the use of such surfaces in different applications, e.g. the additive manufacturing. Four main steps are required to analyse and characterise those surfaces which include new surface representation, surface filtration and decomposition, texture representation methods and finally the calculation of the surface parameters. Recently, the representation, as well as the filtration and decomposition, of freeform surfaces have been investigated and some algorithms have been proposed. This paper, however, shed the light on how to represent the texture of freeform non-Euclidean surfaces before calculating the parameters. A novel model for freeform surface parameterisation is introduced; this new model proposes the use of a projection algorithm before the actual calculation of the parameters. Different projection algorithms have been adopted from the mesh projection techniques found in the field of computer graphics. The results of applying those algorithms to represent the texture of both simulated and bio-engineering surfaces are shown, also a comparison between those algorithms has been carried out. Furthermore, examples of calculating some of the surface parameters for freeform surfaces are given.     

Multi-focus image fusion with Geometrical Sparse Representation

Multi-focus image fusion aims to generate an image with all objects in focus by integrating multiple partially focused images. It is challenging to find an effective focus measure to evaluate the clarity of source images. In this paper, a novel multi-focus image fusion algorithm based on Geometrical Sparse Representation (GSR) over single images is proposed. The main novelty of this work is that it shows the potential of GSR coefficients used for image fusion. Unlike the traditional sparse representation-based (SR) methods, the proposed algorithm does not need to train an overcomplete dictionary and vectorize the signal. In our algorithm, using a single dictionary image, the source images are first represented by geometrical sparse coefficients. Specifically, we employ a weighted GSR model in the sparse coding phase, ensuring the importance of the center pixel. Then, the weighted GSR coefficient is used to measure the activity level of the source image and an average pooling strategy is applied to obtain an initial decision map. Third, the decision map is refined with a simple post-processing. Finally, the fused all-in-focus image is constructed with the refined decision map. Experimental results demonstrate that the proposed method can be competitive with or even superior to the state-of-the-art fusion methods in both subjective and objective comparisons.     

纤维缠绕压力容器几何非线性分析

本文使用有限元软件ANSYS对纤维缠绕压力容器进行了几何非线性分析，并对容器进行了强度以及失效研究。本文压力容器几何非线性分析的结果表明铺层方向是影响纤维缠绕壳体强度的最大因素。     

一种基于改进path opening的道路提取新方法

提出了一种基于改进path opening的高分辨率遥感影像道路提取新方法。为检测曲率较大的道路,构造了新的邻接图;为有效提取有损或有部分遮挡的道路,将path opening算法中的最大路径长度与区域的几何特征相结合,建立了道路区域的判定准则,减少了人工干预。通过多组实验结果证明,该方法对于不完整道路的提取有更好的效果。     

Clifford 代数几何不变量3D医学图像配准的方法

就3D医学图像配准数据量大、计算复杂度高、配准精度低的问题,提出一种基于Clifford代数几何不变量的配准方法,以实现人头颅部3D医学图像配准。提出配准所需的Clifford代数几何不变量及其Clifford代数方程算式,并构造适合于该几何参考轴旋转的Clifford几何旋转算子,利用所求的最大、最小值对应的Clifford几何不变量建立Clifford旋转算子,对浮动影像数据实现几何变换,以达到配准的结果。配准实验中对两个世界著名的3D医学数据集进行了测试,结果表明:该方法计算简单,几何意义直观,配准精度高,执行效率高,并且通过轴线变换不易陷入配准过程的局部极值点。     

Quality- and cost-driven assembly technique selection and geometrical tolerance allocation for mechanical structure assembly

The assembly process planning has been the subject of extensive scientific work, mainly due to the multiple aspects involved from geometrical matters to operational research concerns. However, very few issues about assembly technique selection are addressed. The aim of this paper is to propose a method to select an assembly technique for each joint of a product and to allocate geometrical tolerances accordingly. This is achieved by solving a multi-objective optimization problem to minimize the cost and the non-conformity associated with the assembly plan. The potential benefits of the method are illustrated on a case study representing the assembly of a simple mechanical structure.     

端面模数为标准的斜齿轮测绘

人们习惯以法向模数为标准测绘渐开线斜齿圆柱齿轮，但日常工作中也会遇到以端面模数为标准的渐开线斜齿轮的测绘。本文根据端面模数为标准的渐开线斜齿圆柱齿轮的特点，介绍其几何计算和测绘过程，以及在实践中的应用。