Supervised discrete hashing through similarity learning

Multimedia Tools and Applications - Supervised hashing has achieved better accuracy than unsupervised hashing in many practical applications owing to its use of semantic label information. However,...     

Person re-identification based on multi-appearance model

Person re-identification plays important roles in many practical applications. Due to various human poses, complex backgrounds and similarity of person clothes, person re-identification is still a challenging task. In this paper, we mainly focus on the robust and discriminative appearance feature representation and proposed a novel multi-appearance method for person re-identification. First, we proposed a deep feature fusion method and get the multi-appearance feature by combining two Convolutional Neural Networks. Then, in order to further enhance the representation of the appearance feature, the multi-part model was constructed by combining the whole body and the six body parts. Additionally, we optimized the feature extraction process by adding a pooling layer. Comprehensive and comparative experiments with the state-of-the-art methods over publicly available datasets demonstrated that the proposed method can get promising results.

     

基于多因素改进型PSOSVM算法的中长期负荷预测

中长期负荷预测作为电力规划与调度中的重要一环,其影响因素有着多样性和不确定性等特点.选取支持向量机作为中长期负荷预测的核心算法,筛选多种区域宏观经济因素,利用粒子群(PSO)寻优与循环寻优的改进型算法对支持向量机(SVM)的参数进行优化及负荷预测.仿真结果显示,改进型PSOSVM算法有着较高的预测精度.     

基于特征矢量化的肺结节特征选择算法*

针对肺结节良恶性分类模型中特征选择过程无法避免特征多样性不受破坏的问题,提出一种将肺结节特征矢量化处理的特征选择方法。首先,假设每个肺结节特征都是由数据、类型构成的一个矢量。然后,按照特征类型添加特征到相应的特征子集,并分别利用Relief算法评价特征、特征子集的分类重要性。最后,通过动态阈值的方式筛选得到优化后的特征子集。在150个肺结节样本的分类实验中,采用本文算法所取得的敏感性为94.7%、特异性为93.7%、虚警率为5.2%、受试者工作特性曲线下面积为97.3%。分析表明,本文算法几乎不破坏肺结节特征的多样性,能够显著提高肺结节良恶性分类的准确性。     

基于两级插补机制的NURBS曲线插补器研究

针对非均匀有理B样条（non-uniform rational B-spline,NURBS）曲线高速高精加工过程中计算负载高的问题,提出一种基于两级插补机制的NURBS曲线插补器方案,在主控端执行第一级插补,完成曲线预处理等粗插补工作;在运动控制端执行第二级插补,完成加减速及实时位置输出等精插补工作。同时,设计并开发基于ARM-FPGA架构的实验平台,可用于两级插补器的实验验证,具有一定的现实参考意义。     

高校学生网络评教影响因素研究

抽取4个院系、12个专业,约37 724个原始数据,经过异常值处理后,获得34 022个有效个案。在信度和效度检验的基础上,采取均值相等性的稳健性检验和Tamhane方法进行方差不相等情况下的多重比较,探索了学生网上评教影响因素。研究结果表明,学生年级、学生成绩、教学班级数、教师年龄、授课门数、课程性质等非教学因素显著影响教师教学综合评价。     

基于动态双门限和D-S证据理论的合作频谱感知算法

针对传统能量检测算法门限设置单一且不适应感知信道环境差异的问题,为了提高合作频谱感知结果的可靠性,减小网络开销和融合中心计算的负担,提出了将动态双门限能量检测和D-S证据理论相结合的合作频谱感知算法.在进行本地感知时,该算法可以让认知用户通过双门限能量检测方法直接得到本地判决结果,同时根据信道环境的变化自适应调整双门限值,并在双门限内使用D-S证据理论,感知双门限内未判决结果的信任度函数,将其发往融合中心,这样可以减少融合中心的计算量以及传输带来的网络开销.仿真结果表明,双门限D-S证据理论的融合结果要优于D-S证据理论和OR准则.     

Hydrostatic-pressure-induced depolarization of (Pb1-1.5xLax)(Zr0.80Ti0.20)O3 ferroelectric ceramics

Pulse power energy conversion materials with ultrafast discharge processes and ultrahigh power densities have been widely used in the defense, energy, medical, and mining fields. The pressure-driven depolarization in ferroelectric materials is significant and accounts for the discharge processes. In this study, we focus on pressure-induced depolarization in (Pb_1-1.5xLa_x)(Zr_0.80Ti_0.20)O₃ (PLZT) (x = 0-0.07) ceramics, and their corresponding phase structure, dielectric properties, ferroelectric properties, and thermal depolarization performances. Although a satisfactory pulse power energy conversion performance has been achieved in Pb(Zr_0.95Ti_0.05)O₃ materials, poor temperature stability negatively influences their application. The static charge densities of PLZT (x = 0.04, 0.06) decreased from 29.11 μC/cm² and 31.53 μC/cm² to 19.76 μC/cm² and 6.56 μC/cm² under 400 MPa hydrostatic pressure, respectively, which is attributed to a pressure-driven ferroelectric-antiferroelectric phase structural transition. In particular, the temperature stability of PLZT (x = 0.06) materials is up to 87°C. This study may guide the further development pulse power energy conversion devices.     

The effect of washing process on the cracking of ZrO2 microspheres by internal gelation

ZrO₂ microspheres are widely used as a simulant of UO₂ in the development of nuclear fuel. However, the cracking of ZrO₂ microspheres prepared by internal gelation is still a challenge during drying and sintering processes. To address this issue, we designed and optimized the washing process for obtaining crack-free ZrO₂ microspheres. Through thermogravimetric, infrared, Raman, BET, and SEM analysis, it is shown that the cracking of the microspheres is mainly related to the pores in microspheres. The washing solvent with low surface tension is used to reduce the effect of capillary force on pore shrinkage. Therefore, the optimal washing process was designed as trichloroethylene (TCE)—0.5 M NH₃.H₂O—Propylene glycol methyl ether (PM) and gel microspheres with a high specific surface area of 315.3 m²/g and pore volume of 0.4125 cm³/g were obtained. The characterizations also further showed that when the microspheres were dried and sintered, the water vapor and the decomposition gas of organic matter were completely released from the pores in the microspheres. Our new washing process could be directly extended for preparing crack-free ceramic microspheres by internal gelation.     

High extinction-ratio microstructure fiber based on chalcogenide glasses

Extinction ratio (ER) is one of the important parameters to characterize the polarization-maintaining (PM) performance of the fiber. In this paper, we report the preparation and properties of a novel chalcogenide microstructure fiber with a high ER. We fabricate a preform using a peeled-off extrusion method. The core and cladding material of the fiber are Ge₉As₂₃Se₆₈ and Ge₁₀As₂₂Se₆₈. The preform was drawn into a fiber with an average ER of −17.08 dB. The loss of the fiber is less than 2 dB over 5.20–8.55 μm, and the minimum loss of the fiber is 0.57 dB/m at 6.2 μm. Moreover, a flat mid-infrared supercontinuum spectrum spanning from 1.53 to 12.50 μm is generated by pumping an 18-cm-long PM fiber for the first time.