最优小波尺度空间的图像边缘检测方法

根据图像和检测算子的特性,以相关性为准则,使用遗传算法对图像小波变换的尺度进行选择,从而构成一种自适应的高斯小波尺度空间.融合该空间下不同尺度检测的图像边缘,使得整幅图像的边缘细节丰富清晰,具有更好的抗噪性能.对测试图像使用Canny算法、单尺度、二进尺度和自适应尺度小波进行边缘检测,验证了该算法在去除噪声和准确定位方...     

水分胁迫下高羊茅幼苗的生理适应性

采用不同浓度PEG6000模拟水分胁迫处理6种高羊茅幼苗,测定了胁迫72 h后叶片叶绿素含量、相对电导率、丙二醛含量、超氧化物歧化酶和过氧化物酶活性等抗旱指标,以综合评价6种高羊茅的抗旱能力。综合分析可知,6种高羊茅的苗期抗旱能力最强的是阳光宝贝。     

Denitrification in shallow groundwater in a coastal agricultural area in Japan

In a coastal agricultural area in the central part of Japan (Shizuoka), we found decreasing nitrate concentration with depth in a shallow groundwater, where the depth to water table varied between 0.6 and 1.2 m below ground surface. High nitrate concentrations (5–29 mg N L^–1) were often observed in the upper layer (0–2 m) of the groundwater, but the concentration decreased to less than 1 mg N L^–1 in the deeper layer. Ammonium was scarcely detected, and the concentration of dissolved oxygen was usually low (< 1 mgO₂ L^–1) in the groundwater. Nitrate in the groundwater often had very heavy nitrogen stable isotope ratios (>20{}). There was a negative relationship between nitrogen stable isotope ratio of nitrate and its concentration. When nitrate was injected into the groundwater with acetylene and bromide (a conservative tracer), nitrate concentration decreased to 20% of the initial level within 5 days, accompanied by the increase in nitrite and nitrous oxide concentration and a little change in bromide concentration. These results indicate that microbial denitrification plays a potential role in the decrease of nitrate in shallow groundwater at the study site.     

结合特征适配与拉普拉斯形变的3维人脸重建

目的 数字娱乐产业的发展要求3维人脸重建技术能重建高分辨率3维人脸,并具有较高计算效率和重建准确性。针对这一情况,提出一种基于单幅图像的高分辨率3维人脸重建方法。方法 该方法包含特征适配与拉普拉斯形变两部分。预先用1组3维人脸样本上的3维特征构造可变形模型。给定图像时,从其上自动提取2维特征点,并根据获得问题最优解的必要条件进行特征适配以重建个性化3维特征;然后基于拉普拉斯方法,用该3维特征对一般人脸模型进行变形以获得特定高分辨率3维人脸;最后通过纹理合成获得真实感人脸。结果 用本文方法和已有方法分别进行可变形模型适配和模型变形,本文的特征适配方法具有更快的收敛速度和更高的准确性,拉普拉斯方法具有更小的重建误差。纹理映射后的3维人脸具有很好的视觉效果。结论 本文方法将特征适配与拉普拉斯形变结合起来进行高分辨率3维人脸重建。实验结果表明所提出的方法具有较高的计算效率和准确性,能实现较为理想的高分辨率3维人脸重建。     

Requirements model driven adaption and evolution of Internetware

Today’s software systems need to support complex business operations and processes.The development of the web-based software systems has been pushing up the limits of traditional software engineering methodologies and technologies as they are required to be used and updated almost real-time,so that users can interact and share the same applications over the internet as needed.These applications have to adapt quickly to the diversified and dynamic changing requirements in the physical,technological,economical and social environments.As a consequence,we are expecting a major paradigm shift in software engineering to reflect such changes in computing environment in order to better address the fundamental needs of organisations in this new era.Existing software technologies,such as model driven development,business process engineering,online(re-)configuration,composition and adaptation of managerial functionalities are being repurposed to reduce the time taken for software development by reusing software codes.The ability to dynamically combine contents from numerous web sites and local resources,and the ability to instantly publish services worldwide have opened up entirely new possibilities for software development.In retrospect to the ten years applied research on Internetware,we have witnessed such a paradigm shift,which brings about many changes to the developmental experience of conventional web applications.Several related technologies,such as cloud computing,service computing,cyber-physical systems and social computing,have converged to address this emerging issue with emphasis on different aspects.In this paper,we first outline the requirements that the Internetware software paradigm should meet to excel at web application adaptation;we then propose a requirement model driven method for adaptive and evolutionary applications;and we report our experiences and case studies of applying it to an enterprise information system.Our goal is to provide high-level guidelines to researchers and practitioners to meet the challenges of building adaptive industrial-strength applications with the spectrum of processes,techniques and facilities provided within the Internetware paradigm.     

OpenCOPI: middleware integration for Ubiquitous Computing

In this paper, we present OpenCOPI (Open COntext Platform Integration), a Service-Oriented Architecture-based middleware platform that supports the integration of services provided by distinct sources, ranging from services offered by simple systems to more complex services provided by context-provision middleware. OpenCOPI offers selection and composition mechanisms to, respectively, select and compose services provided by different sources, considering applications of both Quality of Service and Quality of Context requirements. It also offers an adaptation mechanism that enables to adapt the application execution due to service failures, service quality fluctuation and user mobility. OpenCOPI allows the definition of applications in a higher abstraction level by the specification of a semantic workflow that contains abstract activities. This paper illustrates the use of OpenCOPI in an application from the Gas & Oil Industry and it also shows the evaluation of the main mechanisms of OpenCOPI: the service selection, composition, adaptation and workflow execution.     

两种分层方式对温排水三维数值模拟的影响研究

三维数学模型近年来越来越多地被用于研究滨海电厂温排水环境影响问题。海域自由面随潮波动、地形复杂多变,垂向网格的划分方式在一定程度上会影响流场、温度场计算结果。本文以某滨海核电厂为例,采用MIKE 3软件开发斜压三维水动力水温模型,在总分层数即算力保持一致的情况下探究纯sigma分层和垂向混合分层两种分层方式对计算结果的影响。结果表明：两种垂向分层模式对海域潮位的计算精度影响不大,但对有垂向流速梯度的流场刻画会产生一定影响,在近岸海域采用sigma分层模式有利于捕捉垂向流速的变化特征;同时,纯sigma分层方案下,4 ℃最大温升包络面积约为混合分层的4倍,表明不同垂向分层模式可能会对高温升包络面积计算结果产生较为显著的影响。     

孪生导向锚框RPN网络实时目标跟踪

目的 区域推荐网络（region proposal network,RPN）与孪生网络（Siamese）相结合进行视频目标跟踪,显示了较高的准确性。然而,孪生RPN网络（Siamese region proposal network, SiamRPN）目标跟踪器依赖于密集的锚框策略,会产生大量冗余的锚框并影响跟踪的精度和速度。为了解决该问题,本文提出了孪生导向锚框RPN网络（Siamese-guided anchor RPN,Siamese GA-RPN）。方法 Siamese GA-RPN的主要思想是利用语义特征来指导锚框生成。其中导向锚框网络包括位置预测模块和形状预测模块,这两个模块分别利用孪生网络中CNN（convolutional neural network）产生的语义特征预测锚框的位置和长宽尺寸,减少了冗余锚框的产生。然后,进一步设计了特征自适应模块,利用每个锚框的形状信息,通过可变卷积层来修正跟踪目标的原始特征图,降低目标特征与锚框信息的不一致性,提高了目标跟踪的准确性。结果 在3个具有挑战性的视频跟踪基准数据集VOT（video object tracking）2015、VOT2016和VOT2017上进行了跟踪实验,测试了算法在目标快速移动、遮挡和光照等复杂场景下的跟踪性能,并与多种优秀算法在准确性和鲁棒性两个评价指标上进行定量比较。在VOT2015数据集上,本文算法与孪生RPN网络相比,准确性提高了1.72%,鲁棒性提高了5.17%;在VOT2016数据集上,本文算法与孪生RPN网络相比,准确性提高了3.6%,鲁棒性提高了6.6%;在VOT2017数据集上进行实时实验,本文算法表现出了较好的实时跟踪效果。结论 通过孪生导向锚框RPN网络提高了锚框生成的有效性,确保了特征与锚框的一致性,实现了对目标的精确定位,较好地解决了锚框尺寸对目标跟踪精度的影响。在目标尺度发生变化、遮挡、光照条件变化和目标快速运动等复杂场景下仍然表现出了较强的鲁棒性和适应性。     

面向上行流媒体的压缩感知视频流技术前沿

上行流媒体在军民融合领域展现出日益重要的新兴战略价值,压缩感知视频流技术体系在上行流媒体应用中具有前端功耗低、容错性好、适用信号广等独特优势,已成为当前可视通信研究的前沿与热点之一。本文从阐述上行流媒体的应用特征出发,从性能指标、并行分块计算成像、低复杂度视频编码、视频重构和语义质量评价等方面,分析了当前针对压缩感知视频流的基础理论与关键技术,对国内外相关的研究进展进行了探究与比较。面向上行流媒体的压缩感知视频流面临着观测效率难控、码流适配困难和重建质量较低等技术挑战。对压缩感知视频流的技术发展趋势进行展望,未来将通过前端与智能云端的分工协作,突破高效率的视频观测与语义质量导引视频重构等关键技术,进一步开拓压缩感知视频流在上行流媒体应用中的定量优势与演进途径。     

基于双重权重偏差建模的无监督域适应

无监督域适应(Unsupervised Domain Adaptation,UDA)是一类新兴的机器学习范式,其通过对源域知识在无标记目标域上的迁移利用,来促进目标域模型的训练。为建模源域与目标域之间的域分布差异,最大均值差异(Maximum Mean Discrepancy,MMD)建模被广泛应用,其对UDA的性能提升起到了有效的促进作用。然而,这些方法通常忽视了领域之间对应类规模与类分布等结构信息,因为目标域与源域的数据类规模与数据分布通常并非一致。为此,文中提出了一种基于跨域类和数据样本双重加权的无监督域适应模型(Sample weighted and Class weighted based Unsupervised Domain Adaptation Network,SCUDAN)。具体而言,一方面,通过源域类层面的适应性加权来调整源域类权重,以实现源域与目标域之间的类分布对齐;另一方面,通过目标域样本层面的适应性加权来调整目标域样本权重,以实现目标域与源域类中心的对齐。此外,文中还提出了一种CEM(Classification Expectation Maximization)优化算法,以实现对SCUDAN的优化求解。最后,通过对比实验和分析,验证了所提模型和算法的有效性。