基于密集神经网络的灰度图像着色算法

针对在灰度图像着色领域中，传统算法信息提取率不高、着色效果不理想的问题，提出了基于密集神经网络的灰度图像着色算法，以实现改善着色效果，让人眼更好地观察图片信息的目的。利用密集神经网络的信息提取高效性，构建并训练了一个端到端的深度学习模型，对图像中的各类信息及特征进行提取。训练网络时与原图像进行对比，以逐渐减小网络输出结果的信息、分类等各类型的损失。训练完成后，只需向网络输入一张灰度图片，即可生成一张颜色饱满、鲜明逼真的彩色图片。实验结果表明，引入密集网络后，可有效改善着色过程中的漏色、细节信息损失、对比度低等问题，所提算法着色效果较基于VGG网络及U-Net、双流网络结构、残差网络（ResNet）等性能优异的先进着色算法而言取得了显著的改进。     

某工业厂房主体结构安全性评定探讨

结构的安全性是结构防止破坏倒塌的能力,是结构工程重要的质量指标.老工业基地中的厂房主体结构安全性评价,对于厂房加固改造起到了关键性作用.文中通过对某工业厂房主体结构大量现场勘验、隐患排查及检测数据,结合结构验算,从承载能力、损伤等几方面对该老旧工业厂房的安全性进行了评价与分析,并详细介绍了检测鉴定的流程,为以后类似的老旧工业厂房的检测鉴定提供参考.     

基于感知评价的Internet语音终端流量控制

分析了影响因特网语音通信质量的主要因素，根据实时传输控制协议计算延迟、丢包率等网络参数，并利用RBF神经网络将参数映射为语音质量评分，根据评分值的变化动态地调整语音终端的编码速率，在保证一定语音质量的同时，缓解网络拥塞。实验证明了该方法的有效性。     

多层住宅双循环太阳热水系统方案探讨及技术经济分析

提出了用于多层住宅的双循环太阳热水系统设计方案。与每户独立安装的太阳热水器和单循环太阳热水系统相比,双循环太阳热水系统的太阳集热器统一布置设计,可使建筑外观更加整齐,双循环设计使系统具备承压能力高、防冻及防结垢功能等技术特点。根据双循环太阳热水系统初投资较高、使用费低的特点,引入了将初投资、使用寿命期内运行费用、银行存款年利率统一折现为系统运行起始年能源价格的“综合能源价格现值”概念,并使用此概念对双循环太阳热水系统与常规热水系统在相同使用寿命期内的经济性进行了比较分析。     

Frequency Insertion Strategy for Channel Assignment Problem

This paper presents a new heuristic method for quickly finding a good feasible solution to the channel assignment problem (CAP). Like many other greedy-type heuristics for CAP, the proposed method also assigns a frequency to a call, one at a time. Hence, the method requires computational time that increases only linear to the number of calls. However, what distinguishes the method from others is that it starts with a narrow enough frequency band so as to provoke violations of constraints that we need to comply with in order to avoid radio interference. Each violation is then resolved by inserting frequencies at the most appropriate positions so that the band of frequencies expands minimally. An extensive computational experiment using a set of randomly generated problems as well as the Philadelphia benchmark instances shows that the proposed method perform statistically better than existing methods of its kind and even yields optimum solutions to most of Philadelphia benchmark instances among which two cases are reported for the first time ever, in this paper. Won-Young Shin was born in Busan, Korea in 1978. He received B.S. in industrial engineering from Pohang University of Science and Technology (POSTECH) in 2001 and M.S in operation research and applied statistics from POSTECH in 2003. Since 2003 he has been a researcher of Agency for Defense Development (ADD) in Korea. He is interested in optimization of communication system and applied statistics. Soo Y. Chang is an associate professor in the Department of Industrial Engineering at Pohang University of Science and Technology (POSTECH), Pohang, Korea. He teaches linear programming, discrete optimization, network flows and operations research courses. His research interests include mathematical programming and scheduling. He has published in several journals including Discrete Applied Mathematics, Computers and Mathematics with Application, IIE Transactions, International Journal of Production Research, and so on. He is a member of Korean IIE, and ORMSS. Jaewook Lee is an assistant professor in the Department of Industrial Engineering at Pohang University of Science and Technology (POSTECH), Pohang, Korea. He received the B.S. degree in mathematics with honors from Seoul National University, and the Ph.D. degree from Cornell University in applied mathematics in 1993 and 1999, respectively. He is currently an assistant professor in the department of industrial engineering at the Pohang University of Science and Technology (POSTECH). His research interests include nonlinear systems, neural networks, nonlinear optimization, and their applications to data mining and financial engineering. Chi-Hyuck Jun was born in Seoul, Korea in 1954. He received B.S. in mineral and petroleum engineering from Seoul National University in 1977, M.S. in industrial engineering from Korea Advanced Institute of Science and Technology in 1979 and Ph.D. in operations research from University of California, Berkeley, in 1986. Since 1987 he has been with the department of industrial engineering, Pohang University of Science and Technology (POSTECH) and he is now a professor and the department head. He is interested in performance analysis of communication and production systems. He has published in several journals including IIE Transactions, IEEE Transactions, Queueing Systems and Chemometrics and Intelligent Laboratory Systems. He is a member of IEEE, INFORMS and ASQ.     

缓冲区分析在河流沉积相正演研究中的应用

研究区域的沉积相图，是沉积相研究的重要成果。实际工作中，沉积相图的绘制是在综合分析各种地质参数的基础上产生的。应用GIS分析中的缓冲区原理，对影响沉积相的各种参数进行分析研究，并在Microsoft Visual Stdio.NET编程环境中完成缓冲区算法。最后，根据某油田各种实际地质参数数据，应用该方法实现了河道沉积相形成的快速正演模拟，取得了很好效果。     

RF磁控溅射制备钛酸锶钡BST纳米薄膜

钛酸锶钡(BST)薄膜作为一种高K介质材料在微电子和微机电系统等领域具有广阔的应用前景,人们已对BST薄膜的制备工艺技术和介电性能进行了大量的研究。BST纳米薄膜的制备工艺直接影响和决定着薄膜的介电性能(介电常数、漏电流密度、介电强度等)。对RF磁控反应溅射制备BST纳米薄膜的工艺技术进行了综述。从溅射靶的制备、溅射工艺参数的优化、热处理、薄膜组分的控制,及制备工艺对介电性能的影响等方面,对现有研究成果进行了较全面的总结。     

HEGF中CO井筒流动及扩散规律研究

目前,高能气体压裂广泛的应用于低渗透油气田,可以有效清除近井地带由于钻井、射孔和各种措施造成的污染和堵塞,达到油气井增产、注水井增注的目的。但最近一些低渗透油气田却出现了频繁的CO气体中毒事件,给油田及员工带来巨大的损失。针对这一现象,运用现场数据以及物理模拟和数学模拟方法首次通过对CO气体流动、扩散等的综合研究,建立了CO井筒流动及大气扩散模型。同时,也考察了井筒压力、气油比、风速、气体泄放速率、大气稳定度等主要因素对CO气体扩散的影响。通过现场实例计算,证实模型具有较高的准确性。此项工作的完成为建立监控系统提供基础技术依据,同时对合理、有效的开发油气田以及煤层气的开发将具有重要的实际意义。     

Corrosion morphology of AZ91D exposed in an atmospheric environment

The focused ion beam technique was employed to investigate the atmospheric corrosion morphology of AZ91D. It was found that the α matrix of the alloy was preferentially corroded in the areas adjacent to intermetallic phases. The most interesting finding was that the corrosion products in corrosion cavities were pelletlike, which has never been reported before.