国画水墨扩散的LBE计算机模拟

为了满足人们在美工设计中对计算机国画艺术效果仿真的需求,研究和分析了水墨扩散LBE(lattice Boltzmann equation,网格波尔兹曼)算法,提出运用亮度决定水粒子的初始分布,用CMY粒子进行扩散模拟.通过对LBE算法进一步推导而提出疑问,并对扩散及受阻方程进行修正.使用Wintab 1.1压感设备接口完成用户压感数据的读取,借助图像处理函数库IPL98 2.0进行扩散实验数据结构和实验流程设计,实现了国画水墨扩散效果的模拟,证明了该方法的可行性和有效性.     

一种改进的基于视频的车辆检测与识别方法

为了提高基于视频的车辆检测技术在应用中的实时性和准确率,提出了一种应用单目视觉进行车辆检测的方法;首先,提取车道边缘,由车道边缘得到道路区域,根据经验知识在车道区域内确定感兴趣区域,减少车辆检测算法搜索范围;接着基于车辆的对称性特征,阴影和边缘特征对兴趣区域进行过滤,进一步缩小感兴趣区域;最后用离线训练好的AdaBoost分类器对过滤后的图像进行分类识别,检测出动态的车辆;实验结果表明,利用该算法能满足实时性和准确性的要求。     

在线评判在C语言课程设计教学中的应用

本文提出在OnlineJudge系统基础上实现对C语言课程设计进行在线评判的方法,从而减少任课教师的工作负担,提高学生的学习兴趣与主动性,进而提高整个课程的教学质量。     

A best method of demodulation algorithm for UNFSK applied in transportation

The frequency shift of the ultra-narrow frequency shift keying (UNFSK) is far less than the ordinary binary frequency shift keying (FSK) method. Because the two frequencies of UNFSK are extremely close, the traditional demodulation method cannot separate them successfully. This study aims to successfully demodulate the UNFSK signals by applying cycle extension and notch filter. The signal processing procedure is as follows: First, the sampling signals have been made cycle extension to enlarge their spectrum difference between signal 0 and 1; secondly, the cycle extension signals have been processed by the notch filter algorithm and finally, we could get the information by sentencing the filtering result. Finally, the method is applied in transportation. The simulation results proved that the algorithm greatly enhanced UNFSK signal demodulation performance, compared to the traditional demodulation method. There can be 2–3 dB improvement in performance, which is very suitable for UNFSK.     

一种基于图像块的Legendre矩高精度快速算法

针对图像的Legendre正交矩计算量大和矩值求解过程中存在离散近似误差等问题,提出一种新的高精度快速计算图像Legendre矩方法.文中首先提出一种最大块优先分块策略,然后在此基础上,根据图像像素灰度值的取值特征将图像进行分块表示,以每个图像块为单位计算图像的Legendre矩.实验结果表明,与现有的快速算法相比,文中方法在保证矩值高精确的前提下,有效地减少了算术运算的次数,降低了计算复杂度,具有较快的计算速度.     

Evaluating methods of investment project and optimizing models of portfolio selection in fuzzy uncertainty

This paper deals with the problems of both project valuation and portfolio selection under the assumption that the investment capitals and the net cash flows of the projects are fuzzy variables. Using the credibilistic expected value and the credibilistic lower semivariance of fuzzy variables, this paper proposes both the credibilistic return index and the credibilistic risk index, which are measures of investment return and investment risk with annuity form for evaluating single project. Moreover, a composite risk-return index for selecting the optimal investment strategy is also presented. Then, we set up a general project portfolio optimization model with fuzzy returns and two specific models: triangle and interval fuzzy returns. Furthermore, we provide two algorithms: the improved heuristic rules based on genetic algorithm and the traversal algorithm. Finally, two numerical examples are presented to illustrate the efficiency and the effectiveness of these proposed optimization methods.     

基于形态学的蛇形机器人运动步态算法研究

基于蛇形机器人运动形态学,利用扩展Frenet-Serret分析了一种蛇形机器人运动的几何模型,并根据这种几何模型研究了一种确定蛇形机器人的运动步态规律的算法,称之为EFSA(extended frenet-serret algorithm)算法.用这种算法规划了无轮蛇形机器人的基本运动步态,而且重要的是实现了蛇形机器...     

联合纹理和二维阈值分割的减背景方法

为了有效地检测出交通场景中的运动车辆,在背景建模阶段,提出在多个子时间组里建立多个自适应的纹理直方图作为像素的背景描述模型,能更好地适应环境变化。为了充分地描述纹理信息,扩展了LBP(local binary pattern)算子;为了更好地实现减背景方法,采用的二维Otsu阈值分割方法能有效消除噪声的影响;为了解决背景更新滞后的问题,使用帧间差分能更准确地提取出运动车辆。在不同交通场景下的实验比较结果表明了该算法的有效性。     

基于ARM7的无人机监控系统的设计

为了能实时监控无人机设备的状态和提高无人机的安全可靠性,利用ARM7丰富的资源和接口,设计了一套无人机设备监控系统.该系统采用了无线射频模块XT09-SI,利用其长距离特性及高级网络安全等特点,将控制器采集到的信息发送给地面站,并接收地面监控平台发送过来的控制命令.该系统传输的信息不仅包括飞机的姿态及控制命令,还包含传...