融合卷积神经网络与多层感知器的鞍部识别方法

针对传统鞍部识别方法中特征选择困难及未考虑鞍部与其它地形要素的共生关系等问题,利用深度卷积神经网络的特征自学习性能,提出了一种卷积神经网络与多层感知器相结合的混合模型实现DEM数据中的鞍部要素识别.首先设计改进的卷积神经网络模型自动提取鞍部的深度特征,经过Softmax分类器得到候选鞍部点,再运用多层感知器对候选鞍部点的位置进行精细回归,标识出最终的鞍部要素坐标.通过自建的鞍部样本集SADDLE-100训练网络模型,并在三种不同的山地样区进行实验,实验结果表明该方法比其它鞍部识别方法的漏提率减少约50%,正确识别率提高6.7%,在一定程度上避免了人工选择特征造成的鞍部语义信息缺失现象,为DEM中的点状要素识别提供了新的技术途径.     

数字高程模型提取山顶点及空间格局差异

为有效过滤常规山顶点提取方法中的伪点,并探讨不同地貌区山顶点空间分布格局的差异特征,设计了由数字高程模型(DEM)导出等高线的山顶点提取方法。研究以ASTER GDEM为数据源,遴选秦岭中高山区、大巴中山区、低山丘陵区和黄土梁区作为典型地貌样区,采用封闭等高线套合法提取区域山顶点,采用最邻近指数及点间平面距离分析关联格局特征。研究结果表明:秦岭中高山区山顶点最邻近指数为0.76,聚集程度最高;低山丘陵区和大巴中山区分别为0.80、0.82,呈相对聚集;黄土梁区为1.01,接近随机分布,与现实地貌特征相符。研究探索了基于封闭等高线套合法提取山顶的不确定性,突出了山顶点格局特征,对数字地形特征研究有参考意义。     

基于SIFT特征的地貌图像匹配研究

求解地貌图像匹配点对是地貌反求测量过程中的核心问题之一.针对地貌图像的特点,选用基于SIFT 特征图像匹配算法,并根据其算法特点在Vc+ +环境下编写了匹配程序.对地貌模型图像在视角变化、光照不一致等情况下,进行大量实验表明,该方法具有稳定、快速、高效的特点.     

基于视频图像的三维立体生成技术研究

简述了三维数字地图系统开发工作原理以及涉及到的几项关键技术,包括二维图像特征点的提取和三维地貌生成技术。     

砂轮地貌双目图像特征点提取和匹配研究

特征提取和匹配是双目视觉中的关键点和难点。对SIFT和Harris两种特征点提取算法应用于双目砂轮地貌图像的角点检测进行了研究,通过对SIFT特征采用欧式距离进行匹配,对Harris角点采用零均值归一化互相关算法(ZNCC)进行模板匹配,采用RANSAC算法剔除误匹配特征点对。实验结果表明,SIFT算法在双目砂轮地貌图像上应用较Harris算法效果理想。     

基于视频图像的三维立体生成技术研究

简述了三维数字地图系统开发工作原理以及涉及到的几项关键技术，包括二维图像特征点的提取和三维地貌生成技术。     

基于聚类和动态LOD的三维地形建模方法

为提高基于DEM高程数据的三维大地形绘制效率,提出基于K-means聚类和动态LOD的三维地形建模方法。对高程数据集使用K-means聚类分析方法,发现地形数据的自相似性,建立数据采样点与地貌特征分类的自动关联;依据采样点的地貌属性值完成粗化处理,构建不规则三角网模型;采用局部优化算法进行有限域内的三角形分裂以实现三角网的动态更新,结合克里金插值法实现插入点的精确计算,形成LOD层次细节多分辨率模型。实验结果表明,该方法对不同类型地形都获得了较高的数据简化率,提高了三维模型的绘制效率,具有较好的地形特征保持性。     

基于改进SIFT算法的山谷地形影像匹配

针对山谷地形的低空影像中地貌单一且特征不易提取的问题,本文对SIFT算法改进,融合Harris特征提取算法优势,得到一种可用于山谷地形下低空无人机影像特征提取与匹配的算法。算法首先利用Harris算法和SIFT算法分别提取特征点,对两种算法提取的特征点进行合并,然后运用SIFT算法对合并后的特征点进行描述,再利用特征点特征向量的欧氏距离进行粗匹配,最后利用RANSAC算法进行精匹配来提高匹配精度。为了验证该算法的有效性,选用一组山地影像数据进行实验并与SIFT算法进行比较,结果表明:算法有效地提升了山谷地形影像上特征点匹配精度。     

基于面向对象思想的中国地貌形态类型划分

以GTOPO30数据为基础,采用面向对象的分类方法,进行我国地貌形态的自动划分。提取了地形起伏度、地表粗糙度、高程变异系数、坡度变化率、光照晕渲图及平均高程6个地形因子组合成特征影像,并结合《中国及毗邻地区1∶400万地貌图》进行分类。研究结果表明:面向对象思想的遥感分类法可克服传统的基于像元的遥感分类难以利用空间位置信息的缺陷,分类过程更符合人的思维习惯,所分地貌类型更为完整,对提高地貌分类的精度和自动化水平具有重要的意义。     

遥感技术在高放废物处置库场址预选中的应用

运用遥感图像地质解译方法对高放皮物处置库场址预选区内断裂构造特征及其活动性进行了解译，并结合地貌特征和新构造运动的分析对预选区稳定性进行了划分。确定2地区为稳定区。该区符合高放废物处五库场址预选的基本地质条件，可作为进一步进行库址筛选的重点地区。     

An Improved Identity-Based Ring Signcryption Scheme

Abstract

Biometrics can be defined as science by which an individual is identified through analyses of physical data and behavioral characteristics.¹ The measurement of an individual's characteristics quantifies his or her physical, behavioral characteristics. The physically unique characteristics include, but are not limited to, fingerprints, hand or palm geometry, retina or iris scans, voice or facial characteristics, keystroke patterns, and gait. These presumably unique characteristics are used, in turn, to recognize (identify), authenticate, deny, or grant access based on the individual's characteristics.     

Animated agents in K-12 engineering outreach: Preferred agent characteristics across age levels

Animated agents have been found to positively impact student learning and/or perceptions within computer-based learning environments. However, there is little research on the agent characteristics preferred by K-12 students. The main purpose of this study was to examine student preferences for individual pedagogical agents and their preferences for various agent characteristics. Student preferences for the following agent characteristics were assessed using survey methodology: agent gender, age, realism, clothing, personality, speech pace, and tutoring approach. A total of N = 565 students from elementary through high school watched a computer-based multimedia overview of engineering. Four engineering disciplines were introduced by four animated agents: a young female, young male, old female, and old male agent. Immediately after viewing the computer-based overview, students completed surveys assessing preferences for the four agents and for individual agent characteristics. Results indicated that all students preferred agents and specific external agent characteristics that are close to their own external characteristics and favored internal agent characteristics that they felt would promote understanding of the domain. These findings suggest that animated agents used in computer-based K-12 engineering outreach should be close to the student’s external characteristics.     

Comparing narratives of personal and preferred partner characteristics in online dating advertisements

This study explored online personal ads of 294 heterosexual and homosexual men and women in the United States through a qualitative analysis and comparison of participant-generated “personal” and “preferred partner” narratives. Nine characteristics were identified and combined into three overarching categories: physical, lifestyle, and personality characteristics. These three personal and preferred partner characteristics were examined for difference by gender, sexual orientation, age and desired relationship type of the advertisers. Main effects emerged for all four predictors, most notably for age and desired relationship type. Additionally, this study explored the possibility that personal and preferred partner narratives contained similar constellations of characteristics, finding significant correlations on all three variables, lending support for the matching hypothesis in dating partner characteristics.     

Hybrid followee recommendation in microblogging systems

Bug reports are widely employed to facilitate software tasks in software maintenance. Since bug reports are contributed by people, the authorship characteristics of contributors may heavily impact the perfor-mance of resolving software tasks. Poorly written bug reports may delay developers when fixing bugs. However, no in-depth investigation has been conducted over the authorship characteristics. In this study, we first leverage byte-level N-grams to model the authorship characteristics and employ Normalized Simplified Profile Intersection (NSPI) to identify the similarity of the authorship characteristics. Then, we investigate a series of properties related to contributors’ authorship characteristics, including the evolvement over time and the variation among distinct products in open source projects. Moreover, we show how to leverage the authorship characteristics to facilitate a well-known task in software maintenance, namely Bug Report Summarization (BRS). Experiments on open source projects validate that incorporating the authorship characteristics can effectively improve a state-of-the-art method in BRS. Our findings suggest that contributors should retain stable authorship characteristics and the authorship characteristics can assist in resolving software tasks.

     

Renewal processes with Markovian increments

The paper considers the computation of nonstationary characteristics of renewal processes with Markovian increments by analytical-statistical methods and determines the optimal characteristics of these processes by stochastic optimization methods.Translated from Kibernetika, No. 2, pp. 86–90, 100, March–April, 1989.     

一种面向模糊测试的GUI程序空转状态实时检测方法

针对当前Windows下GUI软件模糊测试过程中由于进入空转状态时刻判断不准确导致的测试效率降低的问题,本文利用自然语言处理方法的在函数执行迹的基础上来解决空转状态识别问题.论文首先分析了传统程序分析方法在空转状态判断上遇到的困难,提出了基于Bi-Gram模型以及统计分析的空转状态识别方法.通过Bi-Gram算法将程序函数执行迹转换为概率特征序列;利用空转状态在特征序列中的方差特征将空转状态特征序列从程序特征序列中分离,在此基础上进一步提取空转状态特征并实现空转状态实时检测算法.通过对典型源码与二进制软件程序的实验测试表明,本方法在效率和准确性上较传统方法上更优,能支撑对GUI程序模糊测试的需求.     

An optimum design of robotic food handling by using Burger model

This paper discusses an optimum design approach on robotic food handling by considering the characteristics of viscoelasticity of object. We pick up a traditional Japanese food, “Norimaki” as a typical example with the viscoelastic characteristics. We first show that the dynamic characteristics of Norimaki can be expressed by utilizing the Burger model. After testing the parameter sensitivity, we show an example of the optimum design for determining the combination of the hand stiffness and the operating velocity. We further show that the resultant plastic deformation can be formulated with the exact solution.     

Proposal of a method to analyze 3D deformation/fracture characteristics inside materials based on a stratified matching approach

In the past, deformation/fracture (D/F) characteristics, defined as load-deformation relationships until the materials are fractured, have been analyzed and evaluated on the surface. The D/F characteristics are affected by more than 10,000 micro-scale internal structures like air bubbles (pores), dispersed particles and cracks in 1 mm³; therefore, it is required to analyze nano-scale D/F characteristics inside materials. In this paper, we propose an analysis method by obtaining displacement vectors of dispersed particles from nano-order 3D-CT images. A problem of matching over 10,000 dispersed particles between deformation is solved by a stratified matching.     

Spatial and temporal competition as a two dimensional kinetic Voronoi diagram

An ordinary Voronoi diagram partitions the space into cells with the assumption that all the sites are of the same characteristics. A kinetic site, in addition to its position and hence distance to other sites, is equipped with two more characteristics: an additive constant (such as the time delay to respond to a stimulus) and a multiplicative constant (such as the growth rate or velocity to “catch up”). Under disparate site characteristics, the resulting competition for space becomes unfair: the notion of dominance and conquest ensue. This article reports on the geometry and the topology of kinetic Voronoi regions as pivoted by simple expressions involving dimensionless numbers derived from the site characteristics.     

基于LIBSVM的“就是”句句间关系判别方法

针对使用规则和机器学习方法判别句间关系时出现因机器学习多次迭代而导致规则权值削弱现象,进而导致判别正确率偏低的问题,提出了在规则和机器学习相结合过程中对导入的明显规则特征进行加强处理的方法。首先,抽取依存词汇、语义、句子结构等具有明显规则的特有特征;然后,基于一些句间关系指示词提取普适的特征;其次,将特征写入待输入的数据向量,并且增加一维向量用来存储出现的明显规则特征;最后,运用LIBSVM模型结合规则和机器学习进行实验。实验结果表明,加强后的实验正确率较之加强前平均提高了两个百分点,各句间关系准确率、召回率、F1值整体上都取得了较好的结果,平均值达到了82.02%、88.95%、84.76%。实验思路和方法对研究句子间联系紧密度具有重要价值。