基于深度学习的视频跟踪研究进展综述

近年来深度学习迅猛发展，颠覆了语音识别、图像分类、文本理解等领域的算法设计思路。深度学习因其具备强大的特征提取能力，在图像识别领域的成绩尤为突出。然而深度学习与视频监控领域的结合并不多，由于深度模型具有多层网络结构，算法复杂度大，训练和更新模型时比较耗时，很难满足实时性要求。回顾了深度学习的发展史，介绍了最近10年来国内外深度学习主要模型，论述了基于深度学习的目标跟踪算法，指出了各算法的优缺点，最后对当前该领域存在的问题和发展前景进行了总结和展望。     

检测实验室测量不确定度评定

该文就土的含水率检测测量不确定度评定为例,阐述测量不确定度的评定方法与步骤,旨在与广大读者交流、学习,共同推广测量不确定度在工程实践中的应用,提高检测实验室的检测水平,使测量结果更完整,并与国际接轨。     

TPS分散型控制系统在骤冷风空调装置中的应用

主要介绍了Honeywell公司TPS系统在直纺涤纶长丝用骤冷风空调装置中的应用。重点说明了该系统的硬件配置和一些复杂控制回路的实施和组态。     

FindOut: Finding Outliers in Very Large Datasets

Finding the rare instances or the outliers is important in many KDD (knowledge discovery and data-mining) applications, such as detecting credit card fraud or finding irregularities in gene expressions. Signal-processing techniques have been introduced to transform images for enhancement, filtering, restoration, analysis, and reconstruction. In this paper, we present a new method in which we apply signal-processing techniques to solve important problems in data mining. In particular, we introduce a novel deviation (or outlier) detection approach, termed FindOut, based on wavelet transform. The main idea in FindOut is to remove the clusters from the original data and then identify the outliers. Although previous research showed that such techniques may not be effective because of the nature of the clustering, FindOut can successfully identify outliers from large datasets. Experimental results on very large datasets are presented which show the efficiency and effectiveness of the proposed approach. Received 7 September 2000 / Revised 2 February 2001 / Accepted in revised form 31 May 2001 Correspondence and offprint requests to: A. Zhang, Department of Computer Science and Engineering, State University of New York at Buffalo, Buffalo, NY 14260, USA. Email: azhang@cse.buffalo.eduau     

大朝山水电站碾压混凝土重力坝台阶式溢流面设计

大朝山水电站碾压混凝土重力坝的最大坝高为111m,设计洪水位为899m(500年一遇),此时,表孔下泄的设计流量为9992m3/s,堰顶设计单宽流量为132.7m3(/s·m)。通过多次表孔水工模型试验研究和近三年表孔泄水运行,验证了大朝山水电站碾压混凝土重力坝的表孔采用宽尾墩、台阶式溢流面和戽式消力池联合消能的设计方案是合理可行的,为全段面碾压混凝土重力坝和表孔溢流面的设计开拓了新的思路。     

高频超声波在氧化皮中传播性能标定

对高频超声小在12Cr2MoWVTiB(G102钢）和12Cr1MoV钢钢管运行所产生氧化皮中的传播速度进行了定量标定，并对标定误差进行了统计计算，结果表明，12Cr1MoV钢氧化皮测厚标定点最大偶然误差值为0．050mm,G102钢为0．030mm，并指明了误差来源。     

加强企业文化建设用先进的企业价值观统一员工的思想

通过对企业文化的核心——企业价值观的剖析和理解，提出了“加强企业文化建设，统一全体员工的思想，应着力培育‘价值观’”的看法，对加强企业文化建设工作起到一定的指导作用。     

（1S，2S，5R）-薄荷基取代烯烃的合成

以(L)-薄荷基氯为原料,通过薄荷基氯化镁分别与烯丙基溴、2-甲基-3-氯丙烯、1-溴-3-甲基-2-丁烯在Cul催化下发生偶联反应,首次合成了3-(1S,2S,5R)-薄荷基-1-丙烯,2-甲基-3-(1S,2S,5R)-薄荷基-1-丙烯和2-甲基-4-(1S,2S,5R)-薄荷基-2-丁烯,总收率分别为80.5%、83.1%和79.6%.探讨了影响反应的多种因素,找到了较好的合成条件,并用IR、 1HNMR和元素分析对产物进行了表征.     

腐蚀环境中海工混凝土抗氯离子侵蚀性

模拟我国南方沿海浪溅区和水位变动区的海工钢筋混凝土结构物的腐蚀环境，对掺入粉煤灰和矿渣的混凝土抗氯离子侵蚀性进行了试验。结果表明，在水灰比相同的条件下，加掺和料的混凝土抗氯离子侵蚀性高于普通混凝土。     

Influence of melt superheat on breakup process of close-coupled gas atomization

In close-coupled gas atomization（CCGA）, the influences of melt superheat on breakup process are fundamental to obtain desired or finer powder. Based on a series of Cu atomization experiment under different superheating conditions, the influences of melt superheat on breakup process were studied. Experimental results indicate that as the melt superheat is increased to 150, 200, 250 and 300 K, the mean particle size （D50） decreases consequently to 34.9, 32.3, 30.9 and 19.7 μm. Theoretical analysis reveals that the primary breakup and secondary breakup processes are close coupled, and the melt superheat radically influences the melt properties, and plays a crucial role on governing the filming process of primary breakup and the atomization modes of secondary breakup. There exists a strong nonlinear decrease of contact angle of melt to nozzle orifice wall when the superheat is increased fi＇om 250 K to 300 K, leading to a marked fall of the film thickness formed in primary breakup, and D50 of copper powders is therefore sharply reduced. However, the log-normal distribution feature of particle size has not been substantially improved.