基于主成分分析和分层树集合划分的Huffman算法图像压缩研究

互联网的飞速发展,产生了大量的图像信息.为了减少图片占用的存储空间,提高图像质量,提出了一种将主成分分析（PCA）和分层树集合划分（SPIHT）压缩算法相结合的有损图像压缩算法.首先对图像进行主成分分解,选取主要特征值进行压缩,再利用SPIHT算法将图像分解成不同子带的小波系数进行压缩,对SPIHT压缩系数进行哈夫曼编码,实现图像二级压缩.将本文提出的算法与SPIHT、SPIHT的哈夫曼编码、JEPG2000、PCA压缩算法进行了比较,结果表明本算法较其他压缩算法具有更好的性能,在压缩比相同的情况下能获得更高的PNSR和SSIM.     

旅游目的地形象清晰度及测评方法——以西安为例

基于图像学理论进一步阐释了旅游目的地形象的内涵和形成过程,提出了旅游目的地形象清晰度概念,以及形象清晰度测评指标体系,并以老牌旅游城市西安为案例地,对不同群体旅游者心目中的西安旅游形象清晰度进行了测评.研究表明：（1）旅游目的地形象是旅游者通过视觉、听觉等感觉器官,主动或被动地接收旅游目的地各种旅游信息,并在大脑已有的认知水平下,对所获取的旅游信息进行组合、拼接以及评价形成的目的地心理认知“图像”.（2）旅游者个体之间和群体之间均存在目的地形象清晰度差异.旅游者了解目的地的信息类型越多样、内容越丰富,目的地形象清晰度越高.反之,则清晰度越低.旅游目的地形象清晰度可划分为要素清晰度、维度清晰度和整体清晰度3种类型.（3）旅游目的地形象清晰度与O-D对距离有关.（4）旅游目的地形象清晰度可以用旅游目的地形象描述词汇的频率、多样性指数及词汇网络中心性指标等进行测评.     

一种基于特征分解的图像融合方法

一幅图像可以分解成几何特征不同的纹理部分和卡通部分,基于这两大特征提出了一种图像融合方法.利用卡通和纹理特征的差异,通过学习分别得到卡通字典和纹理字典.在融合过程中,分别利用特定的卡通和纹理字典对源图像的卡通和纹理部分进行融合,融合后的卡通和纹理部分经简单相加得到融合图像.实验结果表明,所提方法是有效的.     

中子检测隐藏爆炸物技术的研究进展

对中子检测隐藏爆炸物技术的基本原理作了较详细的介绍,对几种不同的检测方法进行了比较,分析了当前国内外的研究现状,并指出了未来的发展方向。In this paper, several methods and basic principles of neutron detecting latent explosive are introduced and analyzed in details. The progresses in this field are reviewed and further development direction is also prospected.     

Influence ofΛ Hyperons 1S0 Superfluidity on the Direct Urca Processes with Hyperons

In the framework of the relativistic mean field theory (RMFT), the relativistic energy losses of the direct Urca processes withhyperons (Y-DURCA) for reactionsΛ→p + e +\bar{\nu}_{e} andΞ^- →Λ + e +ν_e arestudied in neutron stars. We calculate the energy gap ofΛ hyperons and investigate the effects of the ¹S₀superfluidity (SF) ofΛ hyperons on the Y-DURCA processes. The calculated results are in line with the recent experimental data\Delta B_ΛΛ∽1.01±0.20^+0.18_-0.11 MeV.The results indicate that the ¹S₀ SF ofΛ hyperons exists in most density ranges of happening the two reactions. The
theoretical cooling curves are in agreement with observation data.     

基于14 MeV中子的烧氚历史诊断模拟研究

针对直接测量16.7 MeV进行烧氚历史诊断所需聚变产额高的情况,模拟研究了利用14 MeV中子与副靶作用产生的非弹伽马进行烧氚历史诊断的情况,计算了几种材料14 MeV中子作用产生的次级伽马能谱以及切伦科夫辐射阈能之上的非弹伽马数目,对副靶材料和厚度进行了选择。计算了14 MeV中子产生的切伦科夫光子时间谱,分析了光电转换器件处伽马、电子以及正电子等噪声信号,分析了气体切伦科夫系统测量统计涨落与聚变中子产额之间的关系,确定了气体切伦科夫系统所适用的最低聚变中子产额,通过测量14 MeV中子与副靶产生的非弹伽马进行烧氚历史诊断较直接测量16.7 MeV伽马可将测量所需聚变中子产额降低2个量级。     

基于模拟退火算法的递归自动阈值分割方法

针对目标图像灰度对比度差的现象,以及对目标对象检测实时性的要求,并考虑到传统的Otsu分割方法在分割图像质量较差以及目标区域小时准确性差的缺点,提出了一种基于模拟退火算法的递归Otsu分割方法。在图像直方图呈双峰的情况下能够准确地找到分割阈值。在成像模糊、光照度较差的情况下此方法仍然可以获得较高的准确度。该方法在保证了检测质量的同时并没有导致运算时间的大幅度提升,有效地保证了处理的实时性。实际应用表明该方法切实可行。     

Energy spectrum of multi-radiation of X-rays in a low energy Mather-type plasma focus device

The multi-radiation of X-rays was investigated with special attention to their energy spectrum in a Mather-type plasma focus device （operated with argon gas）. The analysis is based on the effect of anomalous resistances. To study the energy spectrum, a four-channel diode X-ray spectrometer was used along with a special set of filters. The filters were suitable for detection of medium range X-rays as well as hard X-rays with energy exceeding 30 keV. The results indicate that the anomalous resistivity effect during the post pinch phase may cause multi-radiation of X-rays with a total duration of 300 ± 50 ns. The significant contribution of Cu-Kα was due to the medium range X-rays, nonetheless, hard X-rays with energies greater than 15 keV also participate in the process. The total emitted X-ray energy in the forms of Cu-K and Cu-K/3 was around 0.14 ± 0.02 （J/Sr） and 0.04 ±0.01 （J/Sr）, respectively. The total energy of the emitted hard X-ray （〉 15 keV） was around 0.12± 0.02 （J/Sr）.     

Study on scattering correction in fast neutron tomography at NECTAR facility

Fast neutron tomography has been established as an inspection and detection tool at the NECTAR facility at the FRM-II reactor.Scattered neutrons from the object become a major disturbance and bring in artifacts and deviations in reconstruction results,especially for hydrogenous material object.In this article,an iterative scattering correction method for fast neutron tomography was proposed.In each loop of iteration the scattering component of the projections will be simulated by Monte-Carlo program MCNPX based on the previous reconstruction result and then it will be subtracted from original projections.The differences between scattering components at different perspectives were quantitatively evaluated and an average scattering component image was used for all projections finally.Smooth and uniform slices with more clear edges were obtained and the new reconstructed attenuation coefficients are quite close to the real one compared to the results without scattering correction,in which case the relative error of the reconstructed attenuation coefficients is about 10%–30%.