再生水回用人工景观水体优势藻和水华指示指标的研究

以城市污水处理厂二级出水作为人工景观水体用水,考查了北京5～9月期间藻类的生长情况。研究结果表明,以二级出水作为景观用水的人工水池中,夏季暴发水华的优势藻种类以绿藻为主,主要有小球藻、十字藻、栅藻等,另外还有短时期的裸藻出现,蓝藻一直没有成为优势种属。试验水体的氮磷比为5～15,处于地表水的氮磷比为3～45,可见氮磷比并不是决定水华优势藻的主要因素。对各项指标的监测及分析表明,碳元素的水平才是影响水中优势藻种的关键。试验结果得出水体透明度与藻类生长有明显的相关性,在本试验中当透明度低于0.4m可称已发生水华,可作为水华发生的指示指标。     

浅谈软件过程度量及技术

本文介绍了在软件工程中的度量问题,特别是过程度量的问题.介绍了它的定义和它的用途,并着重介绍了过程度量的方法,如,GQM法、COCOMO模型、功能点模型等.对以后解决软件度量方面的问题有很大帮助.     

基于MPEG-7主色描述符的图像检索

主色描述符作为MPEG-7的一种颜色描述子被广泛的应用于图像检索。通过一种有效的,结构紧凑的,直观的模式,主色描述符描述了一幅图像中的具有代表性的颜色特征。一种新的基于固定数目的MPEG-7主色描述符的图像检索算法被提出。特征提取的过程不再需要阙值的干预以及主色数目固定为8个。直方图相交算法用来衡量特征,从而简化了相似度的计算。实验结果表明,此方法的查准率和查全率高于非固定数目的主色检索方法。     

芝麻显性细胞核雄性不育系内源激素、可溶性糖和淀粉含量变化

植物激素在花药发育过程中具有重要的调控作用。利用酶联免疫检测技术（ELISA）,测定了芝麻显性细胞核雄性不育系叶片和花蕾中生长素（IAA）、脱落酸（ABA）、茉莉酸（JA）以及水杨酸（SA）含量的动态变化,分析了不同内源激素的平衡关系,同时比较了不育和可育株间的可溶性糖和淀粉含量差异。研究结果表明：1）在整个花药发育时期,不育株和可育株中的4种内源激素含量变化趋势明显不同,且含量差异明显;IAA含量的降低及ABA含量的提高可能与芝麻雄性不育的发生密切相关;2）不育株叶片中的IAA含量显著下降,JA和ABA显著上升;3）IAA/ABA、IAA/SA和IAA/JA在不育株和可育株间的变化趋势不一致,且差异很大,表明4种内源激素之间平衡关系受到破坏可能会影响到花蕾的正常生长发育;4）不育系花蕾中的可溶性糖含量、淀粉含量盈余,可能是导致雄性败育的原因之一。     

Correlation between induced embryo toxicity and absorption dose of enriched uranium in testes

ＣｏｒｒｅｌａｔｉｏｎｂｅｔｗｅｅｎｉｎｄｕｃｅｄｅｍｂｒｙｏｔｏｘｉｃｉｔｙａｎｄａｂｓｏｒｐｔｉｏｎｄｏｓｅｏｆｅｎｒｉｃｈｅｄｕｒａｎｉｕｍｉｎｔｅｓｔｅｓＺｈｕＳｈｏｕ－Ｐｅｎｇ（朱寿彭）ａｎｄＬｕｎＭｉｎｇ－Ｙｕｅ（伦明跃）（ＳｕｚｈｏｕＭ...     

基于隐式曲面的三角网格模型等距算法

针对三角网格模型，提出了一种基于隐式曲面的等距算法。该算法首先对三角网格模型进行拓扑重建，然后对顶点进行八叉树采样，由采样点及采样点的单位法矢点来构建隐式曲面，将隐式曲面等距，最后将原模型的顶点投影到等距曲面得到投影点，根据先前建立的拓扑关系，将投影点三角网格化得到等距后的三角网格模型。该算法在一定数值范围内避免了等距模型自交问题，而且等距模型三角网格均匀，质量高。     

块体非晶合金形成机理及成分设计研究述评

简要回顾了块体非晶合金的发展历史，综合评述了块体非晶合金形成的控制因素，从热力学、动力学和结构三方面讨论了块体非晶合金的形成机理，并介绍了块体非晶合金的成分设计准则。指出要使合金形成块体非晶，首先应使合金熔体具有高密度的无规密堆拓扑结构，最大限度地抑制过冷熔体的均质形核及晶核生长；其次，应有适当高的冷却速率，同时减少或消除异质形核。     

GRAPHICAL COMPUTING METHOD ON FOUR-BAR LINKAGE(1)——Inflexion Circle,Focal Axis,Ball''''s Point,Centering-point Curve and Circling-point Curve For Four ISPs,Coupler Curves Based on Ball''''s Points——

By means of programs GTMPAC based- on generalized triangle method,analysis and synthesis of mechanism design in accordance with absolutely graphicalmethod( absolutely germetrical method) are developed.In this paper,we make aspecial study about centering- point curve and circling- point curve and couplercurves based on Ball's points.     

基于OPTICS聚类的差分隐私保护算法的改进

采用聚类算法预先处理个人隐私信息实现差分隐私保护,能够减少直接发布直方图数据带来的噪声累积现象,同时减小了直方图因合并方式不同带来的重构误差。针对DP-DBSCAN差分隐私算法存在对数据参数输入敏感问题,将基于密度聚类的OPTICS算法应用于差分隐私保护中,并提出改进的DP-OPTICS差分隐私保护算法,对稀疏型数据集进行压缩处理,对比采用同方差噪声和异方差噪声两种添加噪声方式,考虑攻击者能够攻破隐私信息的概率,确定隐私参数ε的上界,有效平衡了敏感信息的隐私性和数据的可用性之间的关系。将DP-OPTICS算法和基于OPTICS聚类的差分隐私保护算法、DP-DBSCAN算法进行对比,DP-OPTICS算法在时间消耗上介于其余二者之间,但是在取得相同参数的情况下,聚类的稳定性在三者中最好,因此改进后OP-OPTICS差分隐私保护算法总体上是可行的。     

Model-based early and rapid estimation of COSMIC functional size – An experimental evaluation

ContextFunctional size measurement methods are widely used but have two major shortcomings: they require a complete and detailed knowledge of user requirements, and they involve relatively expensive and lengthy processes.ObjectiveUML is routinely used in the software industry to effectively describe software requirements in an incremental way, so UML models grow in detail and completeness through the requirements analysis phase. Here, we aim at defining the characteristics of increasingly more refined UML requirements models that support increasingly more sophisticated – hence presumably more accurate – size estimation processes.MethodWe consider the COSMIC method and three alternative processes (two of which are proposed in this paper) to estimate COSMIC size measures that can be applied to UML diagrams at progressive stages of the requirements definition phase. Then, we check the accuracy of the estimates by comparing the results obtained on a set of projects to the functional size values obtained with the standard COSMIC method.ResultsOur analysis shows that it is possible to write increasingly more detailed and complete UML models of user requirements that provide the data required by COSMIC size estimation methods, which in turn yield increasingly more accurate size measure estimates of the modeled software. Initial estimates are based on simple models and are obtained quickly and with little effort. The estimates increase their accuracy as models grow in completeness and detail, i.e., as the requirements definition phase progresses.ConclusionDevelopers that use UML for requirements modeling can obtain an early estimation of the application size at the beginning of the development process, when only a very simple UML model has been built for the application, and can obtain increasingly more accurate size estimates while the knowledge of the product increases and UML models are refined accordingly.