Ranking with uncertain labels and its applications

The techniques for image analysis and classification generally consider the image sample labels fixed and without uncertainties. The rank regression problem studied in this paper is based on the training samples with uncertain labels, which often is the case for the manual estimated image labels. A core ranking model is designed first as the bilinear fusing of multiple candidate kernels. Then, the parameters for feature selection and kernel selection are learned simultaneously by maximum a posteriori for given samples and uncertain labels. The provable convergency Expectation Maximization (EM) method is used for inferring these parameters in an iterative manner. The effectiveness of the proposed algorithm is finally validated by the extensive experiments on age ranking task and human tracking task. The popular FG-NET and the large scale Yamaha aging database are used for the age estimation experiments, and our algorithm outperforms those state-of-the-art algorithms ever reported by other interrelated literatures significantly. The experiment result of human tracking task also validates its advantage over conventional linear regression algorithm. A short version of this paper appeared in ICME07.     

平均不等式有趣应用几例

介绍了平均不等式在数学证明中的几例有趣应用。     

JDZ-1型固态存储雨量器降水量误差成因分析

对JDZ-1型固态存储雨量器与虹吸自记雨量计平行观测资料进行了系统分析,分析了两者存在误差的影响因素。对JDZ-1型固态存储雨量器在北方地区使用过程中存在的问题提出了解决方法,以保证和提高JDZ-1型固态存储雨量器资料质量。     

切分轧制技术的发展与应用

简要回顾了近年来切分轧制技术的发展过程，着重介绍了几种典型的切分轧制工艺及其应用，并分析了该技术的发展动向。     

浅谈黑客入侵及服务器安全防护

在网络流行的今天,我们如何更好地保护自己的服务器的安全已是一个更加重要和普及的问题。本文着眼黑客入侵的手段,阐述了服务器安全防护的一些措施。     

NaI多道脉冲幅度分析系统在燃料元件破损监测中的应用研究

介绍了船用堆燃料元件破损监测中较常用的两种方法.通过分析其缺陷,提出了用NaI多道脉冲幅度分析系统测量特征核素(131I、137Cs)的方法,有效避免了监测中的干扰因素的影响,降低了定量监测中的误报率,提高了燃料元件破损监测的效率和置信度.     

The intervalley X6→Γ6, L6 scattering time in GaAs measured by ultrafast pump-probe infrared absorption spectroscopy

A direct measurement of the dynamics of electrons in the X₆ valley for a GaAs crystal by time-resolved absorption spectroscopy is reported for the first time. IR picosecond probe pulses were used to monitor the growth and decay of the population in the X₆ valley subsequent to excitation by a 527 nm pump pulse. The intervalley X₆→Γ₆, L₆ scattering time t_x of 0.70 ± 0.50 ps is determined and the crossection for the X₆→X₇ transition is estimated to be 1.8 × 10⁻¹⁶ cm².     

Fluorimetric determination of monobromobimane and o-phthalaldehyde adducts of gamma-glutamylcysteine and glutathione: application to assay of gamma-glutamylcysteinyl synthetase activity and glutathione concentration in liver

The reversed-phase HPLC separation of fluorescent o-phthalaldehyde (OPA) derivatives has been applied to the assay of hepatic gamma-glutamylcysteine and glutathione (GSH) levels and the enzymes producing these peptides. The method has been compared to the assay using monobromobimane (MB) as the derivatizing agent. The OPA method has the advantage of faster derivatization, the lack of need to adjust the pH, isocratic separation and selectivity for GSH and gamma-glutamylcysteine. The MB method requires pH adjustment following derivatization and gradient elution chromatography. MB is also non-selective, yielding fluorescent derivatives of all biological thiols and more interfering peaks on the chromatogram. MB-based analyses are also approximately sixty times more expensive per sample. MB yields fluorescent degradation products on exposure to light. OPA adducts are stable for up to ten days when stored at -20 degrees C. OPA detection is sensitive to 12.5 pmol in the sample, at a signal-to-noise ratio of 2.5. The two methods correlate well. Hepatic gamma-glutamylcysteine synthetase in the same liver preparation was found to be 4.85 +/- 0.47 nmol min-1 mg-1 protein by the OPA method and 4.42 +/- 0.52 nmol min-1 mg-1 protein by the MB method. GSH concentrations were found to be 90.4 +/- 6.5 nmol/mg protein for the OPA method and 92.5 +/- 3.4 for the MB method.     

温度,压力对漂移室气体增益的影响

研究了温度及气体压力对漂移室（工作于正比区）气体放大倍数的影响，温度和气体压力的变化范围分别是５￣４０℃及９０￣１０５ｋＰａ。用实验数据拟合得到了函数关系并测量了几种饱和度下函数中的参数值。