B/S模式应用系统的一种新型结构及其实现

由于目前Ｂ／Ｓ模式应用系统的缺陷,提出了一种新型的系统结构并从原理上阐述了它的合理性,最后从模块间的通信。处理线程和数据库操作程间的同步和互斥以及系统等方面讲述系统的实现。     

平面内一组线段的可见性

给定平面内一组互不相交的线段,本文讨论其相对于某点的可见性,给出了时间复杂性为O(mlogm)的算法,并证明了这一算法在时间上是最优的.     

Performance evaluation of a dental handpiece in simulation of clinical finishing using a novel 2DOF in vitro apparatus

This paper reports on the performance evaluation of a dental handpiece in simulation of clinical finishing using a novel two-degrees-of-freedom (2DOF) in vitro apparatus. The instrumented apparatus consisted of a two-dimensional computer-controlled coordinate worktable carrying a dental handpiece, a piezoelectric force dynamometer, and a high-speed data acquisition and signal conditioning system for simulating the clinical operations and monitoring the dental finishing processes. The performance of the dental handpiece was experimentally evaluated with respect to rotational speed, torque, and specific finishing energy under the applied clinical finishing conditions. The results show that the rotational speeds of the dental handpiece decreased by increasing either the depth of cut or the feed rate at a constant clinically applied air pressure and water flowrate. They also decreased when increasing both the tangential and normal finishing forces. The specific finishing energy decreased with an increase in either depth of cut or feed rate, while the finishing torque increased as either the depth of cut or the feed rate was increased. Implications of these results were to provide guidance for proper applications of dental handpieces in clinical practice.     

1000BASE—T—在5类铜缆上实现千兆传输

１０００ＢＡＳＥ－Ｔ是一种在５类铜缆上传输１０００Ｍｂｐｓ速度的以太网解决方案。本文介绍了１０００ＢＡＳＥ－Ｔ的基本特性、关键性能参数及其设计。     

金属粉末注射成形粘结剂的发展

概括了金属注射成形粘结剂的功能与发展过程，并对几种典型粘结剂及其脱除工艺进行了适用性与经济性分析。     

DY型风口特性及其在局部空调中的应用

介绍了ＤＹ型风口的结构及射流特性。通过理论分析给出了风口速度衰减的经验公式，介绍了局部空调的定义及ＤＹ型风口在局部空调中的应用实例。     

具有性能感知排序的深度监督哈希用于多标签图像检索

现实生活中的图像大多具有多种标签属性。对于多标签图像,理想情况下检索到的图像应该按照与查询图像相似程度降序排列,即与查询图像共享的标签数量依次递减。然而,大多数哈希算法主要针对单标签图像检索而设计的,而且现有用于多标签图像检索的深度监督哈希算法忽略了哈希码的排序性能且没有充分地利用标签类别信息。针对此问题,提出了一种具有性能感知排序的深度监督哈希方法（deep supervised hashing with performance-aware ranking,PRDH）,它能够有效地感知和优化模型的性能,改善多标签图像检索的效果。在哈希学习部分,设计了一种排序优化损失函数,以改善哈希码的排序性能;同时,还加入了一种空间划分损失函数,将具有不同数量的共享标签的图像划分到相应的汉明空间中;为了充分地利用标签信息,还鲜明地提出将预测标签用于检索阶段的汉明距离计算,并设计了一种用于多标签分类的损失函数,以实现对汉明距离排序的监督与优化。在三个多标签基准数据集上进行的大量检索实验结果表明,PRDH的各项评估指标均优于现有先进的深度哈希方法。