基于SkyEye的嵌入式Linux移植仿真技术研究

在嵌入式系统开发中,经常需要硬件开发板来调测软件,不仅操作繁琐,硬件环境不稳定时问题也很难定位;有时还会遇到软件等待硬件调试,延误开发周期的情况。在PC桌面系统下,探讨了一种基于SkyEye虚拟硬件技术的嵌入式平台仿真技术,介绍如何建立仿真基础环境,编译嵌入式Linux内核,制作根文件系统,编译、配置SkyEye,以S3C6410X为目标机处理器,将Linux内核及文件系统移植到基于SkyEye虚拟的ARM体系架构下,为嵌入式系统的研发探索了一种新的途径。     

基于特征融合的核量化图像分类方法

局部特征和全局特征是图像的两种重要的特征描述,在图像分类时起着至关重要的作用。据此提出一种通过融合全局与局部特征核量化图像分类方法。首先,分析全局特征及局部特征各自优缺点,并对图像进行特征提取;其次,通过核方法将特征映射到适当的高维空间中,来进行码书的获取与量化,并进行特征的融合以更好地对图像进行描述;最后,采用基于直方图交叉核的支持向量机对获取的量化特征进行分类。通过实验证明了所提出的方法的可行性。     

基于贪心核特征提取方法的中期峰值负荷预测

针对中期电力负荷预测, 提出基于贪心核主元回归(GKPCR)、贪心核岭回归(GKRR) 的特征提取建模方法. 通过对核矩阵的稀疏逼近, GKPCR和GKRR两种贪心核特征提取方法旨在寻找特征空间中数据的低维表示, 计算需求低, 适用于大数据集的在线学习. 将所提出的方法应用于不同地区的电力负荷中期峰值预测, 并与现有预测方法进行了比较. 实验结果表明, 在同等条件下, 所提出的方法能有效地改进预测精度, 而且性能更好, 显示了其有效性和应用潜力.

     

Supercritical CO2 impregnation of Radiata pine with organic fungicides: Effect of operating conditions and two-parameters modeling

Supercritical impregnation of Radiata pine with ethyl acetate and decanal using CO₂ as carrier solvent has been studied at pilot plant scale. Radiata pine is one of the most common wood species that is originally from Australia and is widely grown in Spain and Portugal and ethyl acetate and decanal were selected as organic compounds.Some experiences were carried out to obtain the optimal operating conditions for the supercritical impregnation process. Experiments were conducted at pressures of 7–15 MPa, temperatures of 35–50 °C and solvent flow rate between 1.5 and 3.5 kg/h. The results of this study have indicated that the treatment gives much better preservative penetration and retention operating with low pressures (7.5 MPa), low temperatures (close to 35 °C) and moderate CO₂ flow rate (3.5 kg/h) in the selected operating range. Moreover, a simple mathematical model of two adjustable parameters (external mass transfer coefficient and partition coefficient) has demonstrated to fit the experimental impregnation curves with reasonable accuracy (average absolute deviation, 3–10%).     

基于卷积树核的中文微博情感要素识别

情感要素识别是情感分析的关键子任务之一,其目的是识别出文本情感所作用的情感对象。文本情感要素识别属于最细粒度的情感分析,吸引了大量研究者的关注。中文微博由于其语言简短灵活、文本不规范、噪声较大等特点,给中文微博情感分析研究工作带来了新的挑战。目前大部分情感要素识别方法都是基于规则的方法或者基于扁平化特征的统计学习方法,区分噪声的能力不强,性能提升有限。针对中文微博的特点,提出一种基于卷积树核的情感要素识别算法,即首先对句子进行词性标注与依存关系分析,将句子中的名词作为候选情感要素;然后基于两种不同的修剪策略对依存树进行修剪,以获取每个候选情感要素的结构化信息;最后采用卷积树核计算依存树的相似度,并在此基础上识别句子中的情感要素。NLPCC2012和NLPCC2013中文微博情感分析评测任务中的实验验证了该方法的性能,其准确率相比于传统方法有显著提升。     

基于高斯核的SVM的参数选择

基于高斯核的支持向量机应用很广泛,高斯核参数σ的选择对分类器性能影响很大,本文提出了从核函数性质和几何距离角度来选择参数σ,并且利用高斯函数的麦克劳林展开解决了参数σ的优化选择问题。实验结果表明,该方法能较快地确定核函数参数σ,且 SVM 分类效果较好,解决了高斯核参数σ在实际应用中不易确定的问题。     

Algorithm for the Removing Uniformed Motion Blur

Motion deblurring is one of the basic problems inthe field of image processing. This paper summarizes the mathematical basis of the previous work and presents a deblurringmethod that can improve the estimation of the motion blurkernel and obtain a better result than the traditional methods.Experiments show the motion blur kernel loses some important and useful properties during the estimation of the kernel which may cause a bad estimation and increase the ringingartifacts. Considering that the kernel is provided by the motion of the imaging sensor during the exposure and that the kernel shows the trace of the motion, this paper ensures the physical meaning of the kernel such as the continuity and the center of thekernel during the iterative process. By adding a post process to the estimation of the motion blur kernel, we remove some discrete points and make use of the centralizationof the kernel in order to accurate the estimation. The experiment shows the existence of the post process improves the effect of the estimation of the kernel and provides a better result with the clear edges.     

基于“In-VM”思想的内核恶意代码行为分析方法

随着互联网的高速发展,网络安全威胁也越来越严重,针对恶意代码的分析、检测逐渐成为网络安全研究的热点。恶意代码行为分析有助于提取恶意代码特征,是检测恶意代码的前提,但是当前自动化的行为捕获方法存在难以分析内核模块的缺陷,本文针对该缺陷,利用虚拟机的隔离特点,提出了一种基于"In-VM"思想的内核模块恶意行为分析方法,实验表明该方法能够分析内核模块的系统函数调用和内核数据操作行为。     

高斯型模糊润饰图像的模糊核反演算法

为了从模糊润饰图像恢复模糊核,基于对数傅里叶空间图像高斯型模糊润饰前、后的指数黎曼度量具有等距性特点,提出一种高斯型模糊润饰图像的模糊核反演算法.首先将模糊润饰前、后的图像转换到对数傅里叶空间;然后利用黎曼测地距离度量图像高斯型模糊润饰前、后的模糊不变量,从模糊不变量携带的信息中反演恢复出高斯型模糊核.在对高斯模糊、中值模糊、盒式模糊和多步高斯模糊润饰图像进行的实验结果表明,文中算法能够鲁棒地恢复出模糊核.     

Influence of oleoresin constituents fromPinus ponderosa andPinus jeffreyi on growth of mycangial fungi fromDendroctonus ponderosae andDendroctonus jeffreyi

Dendroctonus jeffreyi andD. ponderosae are morphologically similar sympatric species of pine bark beetles over portions of their geographic ranges; however,D. jeffreyi is monophagous onP. jeffreyi whileD. ponderosae is highly polyphagous. Both species carry a species of mycangial fungi that are also very similar in appearance. Growth of the two mycangial fungi and of the fungusLeptographium terebrantis (associated with the polyphagous and non-tree-killingDendroctonus valens) in the presence of oleoresin constituents of host and nonhost conifers was tested by placing individual chemicals on agar growth medium and by growing the cultures in saturated atmospheres of the chemicals. The fungus associated withD. jeffreyi showed greater tolerance for chemical constituents placed on the growth medium than the other two fungi, and growth after three days was enhanced by heptane, the dominant constituent ofP. jeffreyi oleoresin. Growth of all three species of fungi was reduced by the resin constituents when the chemicals were presented as saturated atmospheres. The results suggest that the influence of the tree on growth of the symbiotic fungi of the bark beetles during the initial attack process may be different than after colonization by the beetles is complete. The difference in the responses of the apparently related species of mycangial fungi may provide some new insight into the evolutionary history of these beetle/mycangial fungus/host tree systems.