基于概念聚类的用户兴趣建模方法

Internet资源的指数级增长促进了个性化服务的发展.针对传统的用户兴趣建模方法在准确率和增量处理能力方面的不足,提出了一种新的基于概念聚类的用户兴趣建模方法UIM2C2(User Interest Modeling Method based on Conceptual Clustering).该方法首先通过分析用户访问的历史文档构造后缀树结构,然后选择不同的相似度阈值,以不同的粒度合并基本簇.依据不同阈值条件下合并的基本簇之间的包含关系,生成用户的兴趣层次.UIM2C2方法是针对文档的一个增量式、无监督的概念学习方法,因此用户描述文件可以轻易的获取和更新.最后,通过数据集20NewsGroup上的实验验证了UIM2C2方法在兴趣预测方面的有效性.     

基于退步控制方法的绳系卫星回收张力控制

利用退步控制方法研究了绳系卫星系统TSS(Tethered Satellite System)的平面内回收控制问题,提出了一种新的张力控制律.该控制律设计由两步组成:首先利用退步控制将摆角稳定到目标值,然后通过设计摆角的目标值实现系绳的回收和回收速度的控制.控制率设计了分段连续的目标角度方案和延时环节,可以使TSS回收过程各状态量变化平滑.数值仿真结果表明,该控制律在圆轨道、近圆轨道和大偏心率的椭圆轨道都有很好的控制效果.     

基于GMF方法的捕风一号风速反演分析

全球尺度高时空分辨率海面风场探测是当前全球气象研究及预报预测领域的关注热点之一,传统海面风场探测技术存在测量区域有限,且受天气环境限制明显等问题.基于全球导航卫星系统-反射(GNSS-R)测量技术风速反演原理,以捕风一号1级数据产品为输入,欧洲中期天气预报中心再分析风速数据为参考风速,采用地球物理模型函数(GMF)方法...     

星载SAR海洋场景仿真与反演方法综述

研究SAR海洋目标仿真与反演方法,可提高海洋SAR卫星系统的总体设计能力和海洋图像的应用水平.文章根据星载SAR海洋场景成像机理,对风、浪、内波、锋面等海洋现象的成像模型与反演算法进行了总结,并综述了算法理论现状,分析了SAR海洋场景仿真与反演的研究难点及未来发展趋势,可为我国海洋SAR卫星的总体设计提供参考.     

Nighttime cloud properties retrieval using MODIS and artificial neural networks

In this work a methodology for inferring water cloud macro and microphysical properties from nighttime MODIS imagery is developed. This method is based on the inversion of a theoretical radiative transfer model that simulates the radiances detected in each of the sensor infrared bands. To accomplish this inversion, an operational technique based on Artificial Neural Networks (ANNs) is proposed, whose main characteristic is the ability to retrieve cloud properties much faster than conventional methods. Furthermore, a detailed study of input data is performed to avoid different sources of errors that appear in several MODIS infrared channels. Finally, results of applying the proposed method are compared with in-situ measurements carried out during the DYCOMS-II field experiment.     

长焦距镜面的相位恢复检测技术研究

针对长焦距光学镜面在位检测的难题,引入相位恢复测量原理,提出了一种基于补偿镜的新测量方法以实现镜面的近距离在位检测.对测量原理和测量方案进行了分析,并构建了测量装置.     

Comparative analysis of four different single-frequency PPP models on positioning performance and atmosphere delay retrieval

Single-frequency precise point positioning (SF-PPP) has attracted increasing attention due to its high precision and cost effectiveness. With various strategies to handle the dominant error, i.e., ionosphere delay, the ionosphere-float (IF), ionosphere-free-half (IFH), ionosphere-corrected (IC), and ionosphere-weighted (IW) SF-PPP models are certain to possess different characteristics and performance levels. This study is dedicated to assessing and comparing the four models from model characteristics, positioning performance, and atmosphere delay retrieval. The model comparison shows that IC and IW models are full-rank while IF and IFH models have a rank deficiency of size one that will result in biased estimations, which means the better solvability of IC and IW models. The experiments are carried out based on the 7-day Global Positioning System (GPS) observations collected at 57 global Multi-GNSS Experiment (MGEX) stations and Global Ionosphere Map (GIM) products. The results indicate that the IW model can accelerate SF-PPP convergence and achieve higher positioning accuracy compared to the other three SF-PPP models, especially in kinematic mode. With convergence criteria of 0.25 m in horizontal and 0.5 m in vertical, the east/north/up convergence times of IW model are 0.5/15.0/25.0 min and 0.5/16.0/36.5 min for static and kinematic modes, respectively. The IW model is able to achieve an instantaneous positioning accuracy of 0.28/0.35/0.75 m. In addition, a real kinematic test also demonstrates the best positioning solutions of IW model. Regarding troposphere delay retrieval, the IF, IFH, and IW models obtain a comparable daily accuracy of 3.0 cm on average, while the IC model achieves the worst accuracy of 8.0 cm. For precise ionosphere delay estimation, IW model only needs an average initialization time of 34.3 min, but a longer initialization time of 51.6 min is required for IF model. The daily precision of ionosphere delay estimation for IW model can reach up to 10.8 cm. At the present accuracy of GIM products, it is suggested that the IW model should be adopted for SF-PPP first due to its superior performance in positioning and atmosphere delay retrieval.     

卫星遥感农业监测系统中实例检索算法研究

针对智慧农业中需求获取手段、表达方法等导致智能设计算法复杂度高、可用性低等问题，将卫星遥感技术和网络技术结合实现农作物及其生长环境参数的实时监测，以优化播种机设计的需求特征，并以此为基础，提出基于混合相似度的实例检索模型。在该模型中，设计了多级实例库结构，并以特征间距离相似度、特征间可拓关联度、特征间可拓距测度为组成要素，构造多级混合相似度的层次结构与计算方法，进行多级实例检索。将其应用到铲式播种机的数字化设计系统中，可获得与目标实例最相似的实例。实验结果表明，在实例的特征值数据类型多样化的情况下，此算法能够有效地检索出合适的历史实例，提高检索结果的可用性。     

支持快速数据检索的数据库机系统的研究

本文介绍了一个支持快速数据检索的数据库机TDM系统。TDM系统是一个树结构的关系数据库机系统。它具有紧密耦合式树型多机结构、数据的合理划分与分布、MIMD并行操作方式、多任务的合理调度与分配、多查询优化等特点,这使得数据库操作效率大为增加。该系统并行性高,功能齐备,十分适合于需要快速数据检索的应用场合。文中侧重介绍了TDM系统的功能、体系结构、系统构成及处理方式。     

支持语义的图像检索

如何跨越低层视觉特征和高层语义特征的鸿沟，是目前计算机视觉的一个难点。文中首先在低层视觉特征上提取图像的主要颜色，然后利用语义网络建立低层视觉特征和高层语义特征之间的关联，最后，为了提高检索效率，把相关反馈引进到图像检索系统中。实验证明。该方法可以应用到大量的数据库上，取得了较好的检索准确率。