基于深度图像的人体运动姿态跟踪和识别算法

由于人体运动的复杂性,现有基于低质量深度图像的三维立体姿态跟踪和识别 方法的准确性较低、鲁棒性较差。针对低质量深度图像的人体运动姿态和识别问题,本文设 计了一种基于三步搜索算法的人体运动姿态的跟踪和识别方法。该方法首先对获取的深度信 息进行分析,从而判定人体轮廓;然后通过基于深度图像的骨骼跟踪方法跟踪特定骨骼点, 并采用三步搜索算法进行运动估计,跟踪获取人体运动轨迹;最后利用获取的骨骼点坐标实 现人体运动姿态的识别。实验结果表明,该算法克服光照影响的鲁棒性较强,且能有效地提 高人体运动姿态跟踪与识别的准确性。     

The frequency partitioned block modified filtered-x NLMS with orthogonal correction factors for multichannel Active Noise Control

The Normalized Least Mean Square (NLMS) algorithm with a filtered-x structure (FxNLMS) is a widely used adaptive algorithm for Active Noise Control (ANC) due to its simplicity and ease of implementation. One of the major drawbacks is its slow convergence. A modified filtered-x structure (MFxNLMS) can be used to moderately improve the speed of convergence, but it does not offer a huge improvement. A greater increase in the speed of convergence can be obtained by using the MFxNLMS algorithm with orthogonal correction factors (M-OCF), but the usage of orthogonal correction factors also increases the computational complexity and limits the usage of the M-OCF in massive real-time applications. However, Graphics Processing Units (GPUs) are well known for their potential for highly parallel data processing. Therefore, GPUs seem to be a suitable platform to ameliorate this computational drawback. In this paper, we propose to derive the M-OCF algorithm to a partitioned block-based version in the frequency domain (FPM-OCF) for multichannel ANC systems in order to better exploit the parallel capabilities of the GPUs. The results show improvements in the convergence rate of the FPM-OCF algorithm in comparison to other NLMS-type algorithms and the usefulness of GPU devices for developing versatile, scalable, and low-cost multichannel ANC systems.     

An introduction to the Information Display Measurement Standard 1, the first display measurement standard from the International Committee for Display Metrology

On June 1, 2012, the Information Display Measurement Standard version 1 was released. This is the first standard from the worldwide standard group, the International Committee for Display Metrology, a standards committee within SID, and it culminates several years worth of development from many contributors at various levels. It has 26 top‐level chapters and a total over 450 sections of all parts, including over 200 measurement sections and 8 appendices that cover 133 topics of display metrology, tutorials, technical discussions, and references. Herein is an overview of the Information Display Measurement Standard version 1, including the structure of the document, tips, tricks, and hidden features, how it is maintained, support material, a breakdown of its contents, and finally, a sampling of four of the over 200 measurements. The standard is available as a free download and can be found on the International Committee for Display Metrology home page — http://www.icdm‐sid.org .     

摄影测量中一种非编码点的自动匹配方法

在极限约束法基础上提出了一种非编码点匹配的改进算法,首先采用极限约束法得到候选匹配对应点,然后按编码点位置将图像Delaunay三角化,缩小匹配范用。在对应Delaunay三角形内确定匹配点,若在三角形范围内仍不能唯一的确定匹配点,则通过交比不变性来实现对应点的匹配。本方法不限制标志点的粘贴位置,满足摄影测量中实际应用要求。     

射频宽带功率放大器设计

针对软件无线电台的特点与要求,研制了一款射频宽带功率放大器。首先,判别功率管的稳定性,基于S参数的设计方法,对功放的输入输出匹配电路进行设计和仿真优化。最终使该款功率放大器测试结果达到设计要求。     

煤矿监控图像中脸肤色和模板检测

煤矿监控图像的对比度低、灰度不均匀,使监控图像的处理和识别难度增大.在巷道环境下采集多角度的矿工监控图像,根据脸部肤色区域像素的统计特征,用上下限阈值的方法在HSV颜色空间分割出各种角度的矿工脸部,用半径为2的圆形结构元素进行数学形态学的开启和闭合操作,去除被误分为脸部区域的背景像素;根据脸部灰度分布特征构造了平均脸模板,用相似度函数作为脸部模板匹配的判别函数,检测矿工脸部位置.结果表明,用肤色分割和平均脸部模板匹配方法,可以快速检测出矿工脸部的精确位置.     

特征匹配引擎设计与实现

网络入侵防御系统是维护网络安全的重要工具之一。特征匹配引擎是网络入侵防御系统的计算核心,用于从网络包中搜索出已知网络攻击的特征数据。对于现有的网络入侵防御系统,特征匹配引擎在整个系统中占据很大比例的计算时间。特征匹配属于计算密集型应用,对于底层的计算能力具有很高的要求。提出了一种并行特征匹配算法,并充分利用底层硬件特征,将提出的算法映射到现有的多核处理器上。在IBMSystemx3455上实现的系统具有高达17.2Gbps的处理速度和5万字条的特征字典容量,其处理速度和特征字典容量超过现有文献中的结果,包括FPGA、ASIC、网络处理器以及GPU上的实现。     

基于非精确图匹配的CAD模型搜索方法

为了弥补现有的三维CAD模型搜索方法难以搜索到不同近似程度的相似模型的缺陷,提出一种基于面属性化邻接图非精确匹配的CAD模型搜索方法.首先提取CAD模型中的B-rep信息将CAD模型转化为面属性化邻接图;然后计算目标模型与被搜索模型的面属性化邻接图之间的顶点相容程度矩阵和边相容程度矩阵,并由此建立2个模型相似程度的度量作为选择不同顶点匹配矩阵M的优化目标函数;在对匹配矩阵M进行连续化松弛后,运用Sinkhorn行列交替规范化方法求解匹配优化问题.实验结果表明,采用该方法能够搜索到不同近似程度的相似模型;并且由于避免了具有NP复杂性的精确图匹配过程,检索效率也能满足实际要求.     

改进的遗传算法应用于碎片拼接

碎片拼接实际用途已经越来越广泛,本文讲述传统碎片拼接的方法,引出图形匹配策略,传统的碎片拼接算法大多需要大量繁琐的计算,大大增加了算法的时间复杂度,依据传统算法的不足,改进遗传算法应用于碎片匹配策略,提出可以实现更高效率、更高准确率的碎片拼接算法。