Image recognition of dance moving based on heterogeneous multiprocessor and dynamic image sampling

Motion capture technology has made great strides in computer science and information technology. The design of dance movements requires simulation and dancing of design elements to be critical to overcome the obstacle; 3d aesthetics need to be optimized and cause body rotation problems. Using image analysis technology to establish a mathematical model based on dance motion capture technology, it takes the five main dance movement components as the primary research object. It then proposes the continuity similarity of dance movement. Incorporate composite data obtained from samples during alignment. The proposed system's high tracking accuracy system indicates that it is suitable for various dance training programs based on the game. Video application product video studio software, and programming program, multimedia design of dance movement in the computer, and then realize the effect of dance technology in conjunction with computer technology. Finally, it provides a symbolic reference to dance movement technology research and design.     

基于遗传算法的异构多核多帧任务分配*

讨论一组实时系统的任务在不同性质的处理器上的分配,使得所有任务得以完成并耗费更少的时间,是NP完全问题。建立了新的任务分析模型——异构多核多帧任务模型,并基于遗传算法给出解决方案。实验证明,该模型更为有效地表达了实时系统的性质,获得更高的分配成功率,算法拥有更低的时间复杂度,结果可信。     

动态异构特征融合的水下图像增强算法

针对水下图像细节模糊和色彩失真严重的问题,提出一种基于编码解码结构的动态异构特征融合水下图像增强网络.首先,设计异构特征融合模块,将不同级别与不同层次的特征进行融合,提升网络对细节信息和语义信息的整体感知能力;然后,设计新型特征注意力机制,改进传统通道注意力机制,并将改进后的通道注意力与像素注意力机制加入异构特征融合过程,加强网络提取不同浑浊度像素特征的能力;接着,设计动态特征增强模块,自适应扩展感受野以提升网络对图像畸变景物的适应力和模型转换能力,加强网络对感兴趣区域的学习;最后,设计色彩损失函数,并联合最小化绝对误差损失与结构相似性损失,在保持图像纹理的基础上纠正色偏.实验结果表明,所提出算法可有效提升网络的特征提取能力,降低水下图像的雾度效应,提升图像的清晰度和色彩饱和度.     

Buffer allocation for real-time streaming applications running on heterogeneous multi-processors without back-pressure

The goal of buffer allocation for real-time streaming applications is to minimize total memory consumption, while reserving sufficient space for each data production, without overwriting any live data and guaranteeing the satisfaction of real-time constraints. Previous research has mostly focused on buffer allocation for systems with back-pressure. This paper addresses the problem of buffer allocation for systems without back-pressure. Since systems without back-pressure lack blocking behavior at the side of the producer, buffer allocation requires both best- and worst-case timing analysis.Our contributions are (1) extension of the available dataflow techniques with best-case analysis; (2) the closest common dominator-based and closest common predecessor-based lifetime analysis techniques; (3) techniques to model the initialization behavior and enable token reuse.Our benchmark set includes an MP3 decoder, a WLAN receiver, an LTE receiver and an LTE-Advanced receiver. We consider two key features of LTE-Advanced: (1) carrier aggregation and (2) EPDCCH processing. Through our experiments, we demonstrate that our techniques are effective in handling the complexities of real-world applications. For the LTE-Advanced receiver case study, our techniques enable us to compare buffer allocation required for different scheduling policies with effective impact on architectural decisions. A key insight in this comparison is that our improved techniques show a different scheduling policy to be superior in terms of buffer sizes compared to our previous technique. This dramatically changes the trade-off among different scheduling policies for LTE-Advanced receiver.     

基于图变换的图像压缩采样与分类

提出一种基于图论表示的正交变换基,并在此基础上对图像进行压缩采样与压缩域直接分类.首先,充分利用图像的边缘特性和像素关系,给出一种图像的图论表示方法;然后,通过图Laplacian矩阵的特征值分解得到其特征向量矩阵作为正交变换基,由此得到图像的图变换域稀疏表示;最后,利用随机投影后的压缩采样特征向量直接对分类器进行训练和测试,不仅保持了与原空间相当的分类精度,还大量地减少了训练和测试时间以及计算/存储代价.     

基于动态异构网络的股价预测

股票预测通常被形式化为非线性的时间序列预测任务,但很少有研究者试图通过技术面数据去系统地揭示股票市场内在结构,例如股票上涨或下跌背后的原因可能是业务领域之间的合作或冲突,这些额外信息的增加有助于判断股票的未来趋势。为了充分真实刻画股票市场的交易状态,表达股票之间显式或隐式的关系,提出一种基于动态异构网络的股价预测模型sDHN（stock dynamic heterogeneous network）,综合股票以及所属行业和地域,将其建模为动态异构网络。该模型在网络上引入动态时序特征,创新融合股票节点的四种不同技术层面的相似性图,生成富信息异构图,最后聚合不同元路径中隐含的语义信息生成嵌入,从异构图的角度充分探索股票之间的潜在关联。此外,在三个真实世界的股票数据集上进行了大量实验,所提出的模型准确率比所有基线模型均高出5%~34%,F1-score则高出11.5%~37%,并且在图解释上证明了该方法的有效性。     

基于多处理器的水下潜器定位系统设计

设计了一种对水下潜器实时定位的微处理系统,采用宽带超短基线定位方法,通过在潜器上加装声学传感器来实现对潜器的实时定位.为了实时接收处理声学传感器发出的定位信号,设计了基于多DSP的数字信号处理微系统.该系统采用3片高性能TMS320C6416处理器结合一片FPGA芯片,完成对水下目标的实时数据采集和定位解算功能.试验结果表明:该定位系统对目标的轨迹跟踪误差小于1%斜距误差,具有良好的定位性能.     

基于多属性的动态采样协同过滤推荐算法

针对目前协同过滤推荐算法推荐精度和用户数据在算法中匹配度都不高的问题,提出一种多属性的条件受限波尔兹曼机协同过滤推荐模型（MA-CRBM）。该模型基于实值状态的条件玻尔兹曼机,融合了用户职业和性别属性,充分利用数据集中潜在的评分与未评分信息。在训练过程中,采用动态迭代采样算法对原采样算法进行了改进,克服了训练后期数据采样误差波动太大导致精确度不高的问题。在MovieLens 数据集上的实验结果表明,MA-CRBM模型具有较好的推荐效果,可以有效提升推荐模型的精度和效率。     

基于动态概率抽样的标签噪声过滤方法

在机器学习问题中,数据质量对系统预测的准确性产生了深远的影响。由于信息获取的难度大,人类的认知主观且有限,导致了专家无法准确标记所有样本。而近年来出现的一些概率抽样方法无法避免样本人为划分不合理且主观性较强的问题。针对这一问题,提出一种基于动态概率抽样（DPS）的标签噪声过滤方法,充分考虑各个数据集样本间的差异性,通过统计各个区间内置信度分布频率,分析各个区间内置信度分布信息熵的走势,确定合理阈值。在UCI经典数据集中选取了14个数据集,将所提方法与随机森林（RF）、HARF、MVF、局部概率抽样（LPS）等方法进行了对比实验。实验结果表明,所提出的方法在标签噪声识别和分类泛化上均展示出了较高的能力。     

基于ZigBee的图像采集传输系统的设计与实现

介绍了一种基于ZigBee的图像采集传输无线多媒体传感网络系统的设计与实现。系统采用JPEG压缩摄像头及CC2530无线SoC芯片构成节点,实现了参数设置、图像浏览、拓扑结构显示、数据分析和列表等功能,能够实现图像的采集和无线传输。系统具有自组织组网、多种供电方式、功耗低、稳定性强、可扩展性好等特点。     

基于二分图的两级动态异构网络选择方案

通信技术的发展,使多种接入技术并存的异构网络成为未来通信网络的发展趋势,随着用户业务QoS需求的提高和传输带宽的增加,现有的网络选择算法已经不能满足用户高质量的通信需求。针对异构无线网络频谱资源日益紧缺的问题,提出了由用户端和网络端共同参与的两级动态网络选择方案。该方案包括灰度关联分析法和二分图联合优化匹配算法,通过用户端和网络端的共同决策,算法在有效满足移动用户业务服务质量需求的前提下,优化了系统吞吐量,均衡了网络负载。仿真实验表明,相对传统算法,该方案极大地提高了异构网络频谱资源利用率并降低了用户在无线网络间的切换概率,实现了用户需求和网络资源的合理配置。     

基于异构动态图模型的社交网络节点分类方法

在机器学习领域,与传统的神经网络相比,图神经网络在社交推荐等任务中发挥着越来越重要的作用,但是目前工作中大多数都使用静态图.针对现有静态图神经网络方法难以考虑社交用户动态特性的问题,通过引入动态图模型提出了一种基于异构动态图模型的社交网络节点分类方法.该方法在动态图建模的基础上,通过基于点边交互的节点特征更新机制和基于循环神经网络的时序聚合方法,实现了高效的动态社交网络节点分类.在多个真实数据集上的实验结果表明,提出方法在动态社交网络数据的节点分类方面有较好的效果,对比静态图和动态图的基准方法有显著的提升.     

基于DCBF的流抽样测量算法

为了提高系统的处理效率,减少系统的测量误差,提出了一种基于动态计数型布鲁姆过滤器（Dynamic Counting Bloom Filter,DCBF）的流抽样测量算法。该算法使用基于报文级别的抽样,并通过DCBF进行流查找和统计,且在CBF计数器溢出时动态增加新的CBF。经理论分析和实验表明,该算法不仅提高了系统的运行效率,减少了存储空间的消耗,同时具有准确性和可扩展性,能很好地适用于高速链路的流量测量中。     

Visual feature coding based on heterogeneous structure fusion for image classification

The relationship between visual words and local feature (words structure) or the distribution among images (images structure) is important in feature encoding to approximate the intrinsically discriminative structure of images in the Bag-of-Words (BoW) model. However, in recently most methods, the intrinsic invariance in intra-class images is difficultly captured using words structure or images structure for large variability image classification. To overcome this limitation, we propose a local visual feature coding based on heterogeneous structure fusion (LVFC-HSF) that explores the nonlinear relationship between words structure and images structure in feature space, as follows. First, we utilize high-order topology to describe the dependence of the visual words, and use the distance measurement based on the local feature to represent the distribution of images. Then, we construct the unitedly optimal framework according to the relevance between words structure and images structure to solve the projection matrix of local feature and the weight coefficient, which can exploit the nonlinear relationship of heterogeneous structure to balance their interaction. Finally, we adopt the improving fisher kernel(IFK) to fit the distribution of the projected features for obtaining the image feature. The experimental results on ORL, 15 Scenes, Caltech 101 and Caltech 256 demonstrate that heterogeneous structure fusion significantly enhances the intrinsic structure construction, and consequently improves the classification performance in these data sets.     

基于动态异构的Web信息集成网页分析方法

将动态异构的Web信息资源进行抽取以统一的方式供用户查询和使用,是当前迫切需要解决的问题。介绍了分析相关Web页面的方法和经验,实现了自动提交HTML表单获得所需页面和对页面的信息抽取。最后,实验证明了此方法的有效性。     

基于GPU异构平台的实时CT图像重建系统的研究

摘要：针对采用单CPU CT图像重建时间长,采用CPU集群重建成本及能耗高的问题,本文提出了CPU多线程+GPU的异构重建模型。这种模型采用CPU多线程流水线模式,将整个任务分解为若干个处理阶段,相邻的两个阶段之间以循环缓存连接,上一阶段完成一次计算任务后将数据放到循环缓存里,然后继续下一次的计算任务,下一阶段探测到循环缓存里有数据后,从缓存里取出数据开始计算。各个任务是并行处理任务的,针对某一耗时瓶颈模块再采用GPU并行加速,充分发挥CPU和GPU的计算资源。CPU多线程+GPU模型相对于CPU多线程模型加速16.45倍,相对于串行CT图像重建加速20.5倍以上。将CPU多线程+GPU模型重建的图像与CPU串行程序重建的CT图像比较,数据结果在误差范围内,满足实验设计要求。本文提出的图像重建模型采用成本较低的GPU显卡就实现了性能大幅提升,大大降低了CT图像重建系统的成本及功耗,而成本及功耗的降低会引起CT医疗诊断费用的降低,最终惠及广大病患。     

Miss-aware LLC buffer management strategy based on heterogeneous multi-core

When multiple processor (CPU) cores and a GPU integrated together on the same chip share the last-level cache (LLC), the competition for LLC is more serious. CPU and GPU have different memory access characteristics, so that they have differences in the sensitivity of LLC capacity. For many CPU applications, a reduced share of the LLC could lead to significant performance degradation. On the contrary, GPU applications have high number of concurrent threads and they can tolerate access latency. Taking into account the GPU program memory latency tolerance characteristics, we propose an LLC buffer management strategy (buffer-for-GPU, BFG) for heterogeneous multi-core. A buffer is added on the side of LLC to filtrate streaming requests of GPU. Cache-insensitive GPU messages directly access to buffer instead of accessing to LLC, thereby filtering the GPU request and freeing up the LLC space for the CPU application. Then, for the different characteristics of CPU and GPU applications, an improved LRU replacement taking into account the recent access time and access frequency of the cache block is adopted. The cache misses-aware algorithm dynamically selects the improved LRU or LRU algorithm to fit the current operating state by comparing the miss rate of cache in buffer so that the performance of the system will be improved significantly.

     

大动态范围多曝光图像融合方法

提出一种基于动态范围扩展的多曝光图像多尺度融合方法。讨论融合目标与动态范围分布、细节及颜色的处理策略。在融合规则中提出全局动态范围系数来反映全局照度范围,并指导亮度融合来突出融合结果的动态范围;使用局部对比度和色彩饱和度来反映原始图像的曝光程度,并用于指导色彩融合;使用小波分析作为多尺度融合工具。对算法进行测试并与已有算法的结果进行信息熵和动态范围比较,结果表明该算法对于信息熵和动态范围均有提高。