虹膜识别在肉类食品安全追溯系统中的应用及关键技术研究

为了提高大型动物肉类食品的安全性,在研究肉类食品供应链的安全管理的基础上,将大型动物的虹膜识别技术引入食品供应链管理中,构建基于虹膜识别的大型动物肉类食品安全可追溯系统,达到肉类食品安全信息查询及出现安全问题后的个体溯源的目的。借鉴人眼虹膜识别的相关技术,以牛眼为例对大型动物虹膜识别的相关算法和虹膜定位等关键技术进行研究,运用改进的Sobel算子对图像进行边缘检测;引入二次B样条曲线算法实现了牛眼虹膜的精确定位;分析牛虹膜纹理特征的分布特点,实现了虹膜特征区域的提取及归一化处理。研究成果对大型动物的虹膜识别研究和肉类食品供应链安全管理具有一定的参考意义。     

道路交通紧急救援服务点的优化选址

针对道路交通紧急救援对公平性、效率性、事故并发性的多目标决策需求,提出用于道路交通救援点选址的多目标决策模型。该模型通过最小化最大救援时间满足救援公平性;通过最大化一次覆盖的需求量和最小化救援点到需求点总加权时间满足救援效率;通过最大化备份覆盖的需求量满足事故并发与重大事故的救援。针对多目标决策模型,开发基于模拟退火法的优化解法,结合算例分析,能够获得减少最大救援时间、最大化一次覆盖和备份覆盖需求量的多个目标方向上最优解,实现pareto最优化的成功率均在95%以上。实例计算比较表明,笔者提出的模型与算法可以为决策者提供多目标选址方案。     

中国与美国对外直接投资区位选择的比较研究

基于邓宁的国际生产折衷理论,对比了中美两国投资规模、空间分布和发展阶段,利用灰色关联分析法比较了影响两国投资区位的因子.结果表明,影响美国对外投资的主要因子是人均GNP、CH和CO2的排放量,影响中国的主要因子是直接投资净流入量、航空运输量和CPI;美国对外直接投资属于市场导向型,而中国尚处于转型期,为对外投资第三阶段.     

我国资源型企业对外直接投资区位选择研究——以铁矿企业为例

世界铁矿资源分布不均匀,导致世界各国对铁矿资源的争夺非常激烈。在资源所在国直接投资是获取铁矿资源的主要方式之一,而投资区位选择又直接关系到铁矿企业与企业所在国对外直接投资的成败。分析了我国铁矿企业对外直接投资现状,指出了其存在的问题,探讨了铁矿资源丰富国家的投资环境,对我国铁矿企业对外直接投资风险的影响,并对我国铁矿企业的对外投资区位选择提出了建议。     

Detection of transfluthrin and metofluthrin genotoxicity in the ST cross of the Drosophila Wing Spot Test

In this study, different concentrations of transfluthrin and metofluthrin have been assayed for genotoxicity by using the Wing Spot Test on Drosophila melanogaster. Standard cross was used in the experiment. Third-instar larvae that were trans-heterozygous for the two genetic markers mwh and flr³ were treated at different concentrations (0.0103 mg mL⁻¹, 0.103 mg mL⁻¹ for transfluthrin and 6 μg mL⁻¹, 60 μg mL⁻¹ for metofluthrin) of the test compounds. Feeding ended with pupation of the surviving larvae and the genetic changes induced in somatic cells of the wing’s imaginal discs lead to the formation of mutant clones on the wing blade. Results indicated that two experimental concentrations of transfluthrin and 60 μg mL⁻¹ metofluthrin showed mutagenic and recombinogenic effects in both the marker-heterozygous (MH) flies and the balancer-heterozygous (BH) flies.     

Network design for water quality monitoring of surface freshwaters: a review

To date, many water quality monitoring networks for surface freshwaters have been rather haphazardly designed without a consistent or logical design strategy. Moreover, design practices in recent years indicate a need for cost-effective and logistically adaptable network design approaches. There are many variables that need to be included in a comprehensive yet practical monitoring network: a holistic appraisal of the monitoring objectives, representative sampling locations, suitable sampling frequencies, water quality variable selection, and budgetary and logistical constraints are examples. In order to investigate the factors which affect the development of an effective water quality monitoring network design methodology, a review of past and current approaches is presented.     

Residence and place image

Data are presented for a particular population on its perception of English and Welsh local authority areas in terms of desirability for residence. The method employed permits a distinction between positive and negative judgements as well as a measure of the differing degree to which areas figure in the mind at all. It is established that the areas are seen in terms of a sharply differentiated hierarchy of preference and that there is considerable variation in the extent to which areas are “visible”. In aggregate negative judgements predominate and the most frequently expressed attitude is strong dislike. Places in the south‐west are popular as are some places in the north. London Boroughs are unpopular although in varying degrees. The most favoured place is Cornwall and the least favoured is Wolverhampton. It is argued that perceptions studied in the “direct” way used here are important to complement those employing an “oblique” method in which criteria of choice are assumed or established by social survey and then used to create a profile of preference against which places can be measured and ranked.     

‘The good district’: Newbury as a seed‐bed for success

This study analyses the behaviour of high technology and new firms in the Newbury District of Berkshire in terms of employment, location, genealogy, and local economic impact. It finds that encouraged by good communications networks and a ‘resident electronics establishment’, the direct, as opposed to the indirect, employment consequences are generally encouraging, but not widely replicable. Moreover, the general picture of employment growth masks important differences in labour demand across occupations, particularly with respect to ‘traditional male’ skilled manual workers.     

OPEN CHANNEL INTERPOLATION OF CROSS SECTIONAL PROPERTIES1

ABSTRACT: This paper confirms the use of interpolated data to refine water surface profiles. Sources of error within these computations are due to truncation error, inaccuracies in geometric data, and improper modeling. Confirmation includes the development of an equation that models the effect of data measurement error on the computed water surface profile. A review of interpolation procedures includes a proposed method based upon geometric properties.     

STABILITY AND REACH LENGTH IN WATER SURFACE PROFILE DETERMINATION1

ABSTRACT: Extension of basic step methods of backwater computation to reaches of finite length is examined. Accuracies of commonly accepted hydraulic loss equations under particular water surface profile conditions are compared. Simulation of energy lines within a reach by parabolic curves is found to minimize error provided orientation of the axis of the parabola is selected in accordance with prevailing hydraulic conditions. Theoretical basis for an index reach length beyond which single-step computation from end to end of the reach must be in error is developed. Reduction of this reach length by suitable factors tailored to hydraulic conditions yields a mathematically defined allowable reach length for backwater computation. When reach length does not exceed this allowable reach length, no significant error may be detected. Automatic insertion of synthetic cross sections interpolated between surveyed cross sections when these are inadvertently spaced too far apart enables computation to proceed. This device is error-free for prismatic channls but may introduce error for irregular natural channels. Preliminary trials indicate that results so obtained may be accepted provided the fall in the original reach does not exceed from one to two feet. When this is exceeded, additional cross sections should be surveyed.