一种面向中转换乘的铁路网建模方法研究

铁路旅行依然属于人们首选出行方式,但种类繁多的车型与庞杂路网却经常导致无谓地花费不少钱财和时间。怎样实现旅行方案制定合理化、科学化与兼顾个人独特追求,并且切实地做到最优化评价指标妥善协调,即构成了民众普遍乞盼。结合铁路运营实际,围绕路网建模当中面临的结构体系、关键难题懈决对策等展开深入讨论,给出了问题处置策略、指导原则与技术路线。研究实验表明,通过文中方法对铁路路网进行建模,可以有效地解决铁路路网路径搜索面临的诸多难题,保证了数据完备性、有效性以及路径规划的实时性与高效性。     

基于OWL的成语典故本体构建研究

本体是共享概念模型的明确的形式化规范说明,作为知识表示和知识共享的一种方法,本体是目前信息处理领域研究的热点。基于本体论的思想,利用骨架法构建了成语典故本体,并用OWL语言对成语典故本体进行形式化描述。详细介绍了成语典故本体的目的和使用范围、知识采集提炼及OWL描述,OWL描述分别从类、子类、属性、个体及关系几个方面进行了详细分析,为成语典故相关知识的查询奠定基础。通过成语典故本体的构建可有效对成语典故进行智能检索,是本体技术在中国传统文化中应用的尝试。     

基于非对称打包和FSVM的图像检索

在图像检索的相关反馈中,引入支持向量机分类方法虽可以提升图像的检索性能,但是传统的支持向量机存在正样本数少、样本非对称、过学习和弱实时性的局限。针对上述问题,提出了一种基于非对称打包的FSVM算法。该算法首先对负样本进行非对称打包处理,最后结合模糊理论与SVM实现图像检索。Corel图片集上的实验表明,当正样本数较小时,该新算法的平均查准率-查全率要优于已有算法。     

基于视觉面距离的三维模型检索

研究计算机视觉图像优化问题,针对目前三维模型检索较低的的检索精度,为改善检索结果与人视觉上保持一致,减少计算量,提出一种新的基于视觉面距离的模型检索算法。利用模型的整体特征;根据表层视觉特性构建视觉面距离模型将三维模型映射到二维模型;由于二维傅立叶算法不仅对噪音具有很好的鲁棒性,而且对几何变换具有不变性。采用二维傅立叶变换提取二维模型特征向量对特征进行融合,从而有效地评估三维模型之间的相似度。进行仿真分析,仿真结果表明,改进方法较传统其他三维模型检索有更高的精度,并与人视觉的一致性。     

基于本体的检索方法在数字期刊中的应用研究

基于关键词处理的传统检索技术会在检索过程中遗漏大量与检索概念相关或同义的内容。本文在本体基础上重点研究语义相似度算法及相应的语义扩展算法,在此基础上将模型应用于数字期刊的信息检索中,以提高查准率和查全率。     

一种基于DotLucene搜索引擎的自动答疑系统

比较了全文检索方法与常规的数据库内搜索的不同之处,提出应用DotLucene技术实现知识库站内中文全文检索的方法,事实表明,索引和查找的时空效率都很高。     

基于MPEG-7主色描述符的图像检索

主色描述符作为MPEG-7的一种颜色描述子被广泛的应用于图像检索。通过一种有效的,结构紧凑的,直观的模式,主色描述符描述了一幅图像中的具有代表性的颜色特征。一种新的基于固定数目的MPEG-7主色描述符的图像检索算法被提出。特征提取的过程不再需要阙值的干预以及主色数目固定为8个。直方图相交算法用来衡量特征,从而简化了相似度的计算。实验结果表明,此方法的查准率和查全率高于非固定数目的主色检索方法。     

Image retrieval model based on weighted visual features determined by relevance feedback

An accurate and rapid method is required to retrieve the overwhelming majority of digital images. To date, image retrieval methods include content-based retrieval and keyword-based retrieval, the former utilizing visual features such as color and brightness, and the latter utilizing keywords that describe the image. However, the effectiveness of these methods in providing the exact images the user wants has been under scrutiny. Hence, many researchers have been working on relevance feedback, a process in which responses from the user are given as feedback during the retrieval session in order to define a user’s need and provide an improved result. Methods that employ relevance feedback, however, do have drawbacks because several pieces of feedback are necessary to produce an appropriate result, and the feedback information cannot be reused. In this paper, a novel retrieval model is proposed, which annotates an image with keywords and modifies the confidence level of the keywords in response to the user’s feedback. In the proposed model, not only the images that have been given feedback, but also other images with visual features similar to the features used to distinguish the positive images are subjected to confidence modification. This allows for modification of a large number of images with relatively little feedback, ultimately leading to faster and more accurate retrieval results. An experiment was performed to verify the effectiveness of the proposed model, and the result demonstrated a rapid increase in recall and precision using the same amount of feedback.     

HD/SHD still image database system and image distribution in broadband network applications

This paper focuses on a practical design for an efficient scalable image database and retrieval system over broadband networks. It describes a concrete solution for the implementation of HD/SHD (high-definition/super-high-definition) still image retrieval services which can be used in different applications. The structure of the complete system, consisting of a directory server, an image server, and MMI (man-machine interface) devices, has been presented, detailing each element and their corresponding functions. The desired HD/SHD image is displayed on the HD-PDP (plasma display panel) with the aid of image matching. The proposed system generates image index automatically, eliminating special skills in preparing index images and crucially reducing the processing time (from 35 min to 110 s), and does not use keywords. It has been also shown that these indices can be used for quite accurate image retrieval, i.e., the system provides high precision rates (currently up to 98%) to the user, eliminating troubles encountered in the image retrieval operations due to limitation on the user’s age, culture, knowledge, and languages.The broadband IP networks currently have a number of issues from the viewpoint of practical system operations, and so the requirements and issues needed for the networks are discussed from the viewpoint of in-service performance, differentiation among different types of services, secure connections, and so on, focusing on handling of HD/SHD still images.     

On technology for functional requirement-based reference design retrieval in engineering knowledge management

Engineering design is a knowledge-intensive process, and includes conceptual design, detailed design, engineering analysis, assembly design, process design, and performance evaluation. Each of these tasks involves various aspects of technical knowledge and experience. Whether this technical knowledge and experience can be effectively shared is key to increasing product development capability and quality, and also to reducing the duration and cost of the development cycle. Consequently, providing engineering designers various query methods for retrieving engineering knowledge is one of the most important tasks in engineering knowledge management.The study develops a technology for functional requirement-based reference design retrieval as a decision support mechanism, which can assist engineering designers to retrieve relevant design and associated knowledge for reference in conducting functional requirements of a product. This study involves the following tasks: (i) designing a functional requirement-based reference design retrieval process, (ii) developing techniques related to the technology for functional requirement-based reference design retrieval, and (iii) implementing a functional requirement-based reference design retrieval mechanism. The retrieval process includes the steps of functional requirement-based query, case searching and matching, and case ranking. The technology involves (i) a structured query model for functional requirements, (ii) an index structures for historical design cases, (iii) functional requirement-based case searching and matching mechanisms, (iv) a functional requirement-based case ranking mechanism, and (v) a case-based representation of designed entities. Finally, the experimental example with indexing and retrieving similar designed entities is conducted to demonstrate the proposed techniques worked efficiently.