FTO客体3m闪光照相的Monte Carlo研究

 研究了客体模型FTO的闪光照相系统X光输运过程，给出了直穿照射量、散射照射量、直散比、直穿照射量能谱、散射照射量能谱、直穿X光通量能谱和散射X光通量能谱在记录平面的空间分布。结果表明：后锥是照射量散射成分的主要来源，后锥照射量占总散射量97%；后锥也是造成散射的空间分布不均匀的主要器件，这一不均匀性高达58%。照相系统的最小直散比非常小，表明锥造成的散射已经严重地淹没了直穿（轫致辐射）信号。计算中使用高空间分辨率记录法进行分点，合成图像对吸收系数的复原结果与国外报道的结果相符。     

Accelerating the calculations of binary detour phase method by integrating both CUDA and Matlab programming for GPU's parallel computations

The availability of state of art GPGPU cards can play a promising role in accelerating the calculations of computer generated holograms (CGH) where one of the main problems of generating such holograms is the need for massive amount of calculations. Both CUDA and Matlab can be used alone to fulfill this purpose but using both at the same time gives a number of benefits. This paper is dedicated to present a performance study of applying Fermi-Architecture CUDA-enabled GPGPU card for speeding up the calculations of binary detour phase holograms using both Matlab and CUDA programming.     

20 Hz~10 kHz光纤水听器相移灵敏度校准

利用相位生成载波解调技术精确测量光纤水听器的光相移量,在20 Hz~10 kHz频率范围实现了光纤水听器探头相移灵敏度的校准.20 Hz~1.25 kHz频段采用驻波管比较法进行校准,1.25 kHz~10 kHz频段采用自由场脉冲比较法进行校准.利用本文建立的校准系统,对TMD 35#光纤水听器的相移灵敏度进行校准,校准结果表明,两种方法测得的相移灵敏度具有很好的一致性,在1.25 kHz频率点的相移灵敏度值偏差为0.8 dB.不确定度分析表明,该校准系统的扩展不确定度(k=2)为0.9 dB.     

Holographic watermarking for authentication of cut images

A watermarking technique, with a Computer Generated Hologram (CGH) coding system of the mark, is introduced and tested. The CGH watermarking can be used to authenticate parts of the original image. The hologram of the mark is embedded in the spatial domain by a blind additive embedding technique. The use of holography allows authenticating cuts of the original image, is the major novelty of this paper. The proposed methodology is characterized as an authentication technique, since it does not rely on the original image to decide whether the watermarked image has been altered or not and at the same time it is able to detect and localize any possible malicious change. Asymmetric cryptography is used to hide the hash information in an unambiguous way (non-repudiation property).     

Stochastic regression model with heteroscedastic disturbance

This paper discusses some properties of stochastic regression model with continuous form of heteroscedastic disturbance. The strong consistency and asymptotic normality of a generalized weighted least squares estimate will be investigated under certain conditions on the stochastic regressors and errors. More, the linear hypothesis testing problem also be discussed and an example to be demonstrated to reestablish the results of Cheng and Chang (1990, Tech. Report, National Tsing Hua University).Supported by the National Science Council Grant No. 810208M763 at National Tsing Hua University.     

条件交半格中的相对极大滤子

引入偏序集的相对极大滤子的概念,证明在任意条件交半格中一个滤子是相对极大滤子当且仅当它是滤子格的完全交不可约元.一个格是分配的当且仅当每一个相对极大滤子都是素滤子.随后研究了Heyting代数中相对极大滤子的刻画,最后定义和研究了完全并既约生成格.     

基于显示、生成偏好的序贯模糊选择函数的研究

本文分别在普通情况以及模糊情况下,对显示偏好、生成偏好合理化的序贯选择函数进行了一定的研究。首先,研究了该序贯合理的选择函数的选择结果与选择顺序无关的条件,其次,研究了选择函数与该序贯合理的选择函数之间的关系。     

Patterning of Aluminium thin film on polyethylene terephthalate by multi-beam picosecond laser

High speed patterning of a 30 nm thick Aluminium thin film on a flexible Polyethylene Terephthalate substrate was demonstrated with the aid of Computer Generated Holograms (CGH׳s) applied to a phase only Spatial Light Modulator. Low fluence picosecond laser pulses minimise thermal damage to the sensitive substrate and thus clean, single and multi-beam, front side thin film removal is achieved with good edge quality. Interestingly, rear side ablation shows significant Al film delamination. Measured front and rear side ablation thresholds were F_th=0.20±0.01 J cm⁻² and F_th=0.15±0.01 J cm⁻² respectively. With laser repetition rate of 200 kHz and 8 diffractive spots, a film removal rate of R>0.5 cm² s⁻¹ was demonstrated during patterning with a fixed CGH and 5 W average laser power. The effective laser repetition rate was f_eff~1.3 MHz. The application of 30 stored CGH׳s switching up to 10 Hz was also synchronised with motion control, allowing dynamic large area multi-beam patterning which however, slows micro-fabrication.     

Classification of dicoverings

The dicoverings of a “well pointed” d-space are classified as quotients of the universal dicovering space under congruence relations. We prove that the subcategory of d-spaces generated by the subcategory of directed cubes is equal to the category generated by the interval and the directed interval. Similarly, the category of topological spaces generated by simplices may be generated by the interval.     

基于迈克尔逊干涉的光纤弯曲传感器

设计并研制了一种新型的基于迈克尔逊干涉的光纤弯曲传感器.该光纤弯曲传感器由两根光纤组成的光纤带固定于弹性弯曲结构上来产生迈克尔逊干涉信号,用光纤折射率匹配液填充微型腔,并施加音频振荡信号对相位干涉信号实现低频调制.采用相位生成载波技术对相位干涉信号进行调制和解调,实现了对曲率变化高精确度的检测.实验测试了该光纤弯曲传感器的相位干涉信号与光纤带弯曲曲率的关系,并与理论分析对比.结果表明该光纤弯曲传感器具有43.96 rad/m－1的灵敏度及0.004 m－1的高分辨率.