Ultra-lightweight CNN design based on neural architecture search and knowledge distillation: A novel method to build the automatic recognition model of space target ISAR images

In this paper,a novel method of ultra-lightweight convolution neural network(CNN)design based on neural architecture search(NAS)and knowledge distillation(KD)is proposed.It can realize the automatic construction of the space target inverse synthetic aperture radar(ISAR)image recognition model with ultra-lightweight and high accuracy.This method introduces the NAS method into the radar image recognition for the first time,which solves the time-consuming and labor-consuming problems in the artificial design of the space target ISAR image automatic recognition model(STIIARM).On this basis,the NAS model’s knowledge is transferred to the student model with lower computational complexity by the flow of the solution procedure(FSP)distillation method.Thus,the decline of recognition accuracy caused by the direct compression of model structural parameters can be effectively avoided,and the ultralightweight STIIARM can be obtained.In the method,the Inverted Linear Bottleneck(ILB)and Inverted Residual Block(IRB)are firstly taken as each block’s basic structure in CNN.And the expansion ratio,output filter size,number of IRBs,and convolution kernel size are set as the search parameters to construct a hierarchical decomposition search space.Then,the recognition accuracy and computational complexity are taken as the objective function and constraint conditions,respectively,and the global optimization model of the CNN architecture search is established.Next,the simulated annealing(SA)algorithm is used as the search strategy to search out the lightweight and high accuracy STIIARM directly.After that,based on the three principles of similar block structure,the same corresponding channel number,and the minimum computational complexity,the more lightweight student model is designed,and the FSP matrix pairing between the NAS model and student model is completed.Finally,by minimizing the loss between the FSP matrix pairs of the NAS model and student model,the student model’s weight adjustment is completed.Thus the ultra-lightweight and high accuracy STIIARM is obtained.The proposed method’s effectiveness is verified by the simulation experiments on the ISAR image dataset of five types of space targets.

World-Wide Semantic Web of Agriculture Knowledge

The rapid increase in the publication of knowledge bases as linked open data （LOD） warrants serious consideration from all concerned, as this phenomenon will potentially scale exponentially. This paper will briefly describe the evolution of the LOD, the emerging world-wide semantic web （WWSW）, and explore the scalability and performance features Of the service oriented architecture that forms the foundation of the semantic technology platform developed at MIMOS Bhd., for addressing the challenges posed by the intelligent future internet. This paper＂ concludes with a review of the current status of the agriculture linked open data.

Knowledge Map Analysis of the Domestic Rural Landscape Research Situation From 1998 to 2018 Based on CNKI

Urban-rural integration and beautiful rural construction have become an important way to improve the ecological environment of urban and rural areas,improve the quality of people’s lives and seek spiritual civilization.Rural landscape has become the research focus of scholars and elites in colleges and universities.In order to better clarify the context and hot spots of domestic rural landscape research,the study took the Chinese journal full-text database(CNKI)as the data source,took 2,079 rural landscape research papers published in the 20 years from 1998 to 2018 as a sample,and used CiteSpace software to carry out a literature econometric analysis of the study samples.According to the analysis results,the main research hotspots of rural landscape were combed and summarized,and the future trend of rural landscape research were predicted,in order to provide reference for researchers to carry out relevant research.

Knowledge Based Consolidation of UML Diagrams for Creation of Virtual Enterprise

In this paper we address the problem related to determination of the most suitable candidates for an M&A (Merger &Acquisition) scenario of Banks/Financial Institutions. During the pre-merger period of an M&A, a number of candidates may be available to undergo the Merger/Acquisition, but all of them may not be suitable. The normal practice is to carry out a due diligence exercise to identify the candidates that should lead to optimum increase in shareholder value and customer satisfaction, post-merger. The due diligence ought to be able to determine those candidates that are unsuitable for merger, those candidates that are relatively suitable, and those that are most suitable. Towards achieving the above objective, we propose a Fuzzy Data Mining Framework wherein Fuzzy Cluster Analysis concept is used for advisability of merger of two banks and other Financial Institutions. Subsequently, we propose orchestration/composition of business processes of two banks into consolidated business process during Merger &Acquisition (M&A) scenario. Our paper discusses modeling of individual business process with UML, and the consolidation of the individual business process models by means of our proposed Knowledge Based approach.

Three-Tier Knowledge Management System Based on .NET

Three-tier knowledge management system based on .NET architecture is designed according to requirement specification, characteristics of and relationship between enterprise electronic archives and knowledge management. This system using three-tier design based on factory pattern has good encapsulation and portability, with clearer and more concise structure. It degrades the costs of system development and maintenance and upgrades system’s high reusability and development efficiency.

Research Progress of Garden Microclimate in Recent 15 Years in China:Knowledge Graph Analysis Based on Citespace

Taking CNKI database as the data source and measuring visual map analysis tool Citespace as auxiliary tool,domestic 843 articles in recent 15 years are analyzed,and visualization maps such as time line analysis,institutional cooperation network analysis,author collaboration network analysis,keyword co-occurrence,and emergent words analysis are drawn.Combined with the literature content analysis,four hot spots in the research field of landscape architecture microclimate in China are obtained,namely ENVI-MET,comfort,design strategy and urban green space.The research trend is thermal comfort,human body comfort and winter city.The research results can provide reference for the research on domestic garden microclimate.

SOA Information Engineering System Based on Knowledge Grid Applications in Smart Grids

Smart grids have the characteristics of being observable,controllable,adaptive,self-healing,embedded independent processing,and real-time analysis.With the development of smart grids,constructing a grid to cover global,unified information systems,which should be adapted to fulf ill the requirements of the characteristics,is essential.This paper presents an service-oriented architecture(SOA)for smart grid information-engineering systems based on knowledge grid,which could form as a service-oriented architecture through business,technology and management;it would extract potentially valuable information from the massive amount of information on the generation side,the grid side,and the electricity side,then share the useful information to improve availability,security and stability.

关于风景园林专业学位与知识体系的思考

【目的】中国风景园林教育事业进入新的发展阶段,在专业学位背景下,风景园林知识体系亟待梳理与反思。【方法/过程】基于专业学位教育特点,剖析国家政策、社会需求,重新系统认知风景园林专业学位:从培养目标、知识类型、知识层级与知识结构4个方面探讨风景园林专业学位与知识体系的内涵及发展方向。现代风景园林专业人才培养的核心在于对实践智慧与能力的培养,专业知识类型势必从过去以经验知识为主体转变为以科学知识与规律为主体。科学、艺术与技术3种类型知识的融合是专业发展与人才培养的基础。学士、硕士与博士三级人才培养模式体现出风景园林专业知识的层级属性,据此处理好专业知识体系中内涵与外延的结构关系是推动专业发展的根本问题。【结果/结论】面对风景园林专业乃至思维范式的革新,应按照专业知识生产与组构的特定规律,探索风景园林专业未来可持续发展的有效路径。

复杂环境下电磁目标信号认知处理架构与应用研究

针对现代电磁频谱设备或终端具有网络化、捷变性、多功能、自适应、种类复杂多样和使用环境复杂多变等特性给认知电磁频谱战系统的信号分析与处理带来的巨大挑战的问题,提出了一种结合人工智能、适合复杂电磁环境下电磁目标信号认知处理与分析的架构。在该架构下电磁目标信号分析与处理被分为目标信号参数估计、目标信号类型分类、辐射源个体识别和目标信号的特征建库与知识图谱[1-3]的构建等4个层次,可以很好适应闭集和开集电磁目标信号的分析与处理;此外,在此架构下提出了未知辐射源辨识、跨模态辐射源智能识别和电磁辐射源的个体识别的算法框架,在实测数据集上的验证实验表明,该算法在闭集和开集电磁信号分析中有效可行。

面向技艺传承的传统建筑古籍的知识组织——以《营造法式》大木作制度为例

传统建筑古籍是古代建造技术的重要载体,对其技艺与经验进行语义关联和组织对古建筑建造技术的传承与利用具有重要意义。文章以经典建筑古籍《营造法式》卷四、卷五所述“大木作制度”工种的文字信息为原始语料,以大木作营造技艺为对象,在《古建筑类文物元数据规范》的基础上,进行“宋式大木作营造技艺”本体的构建、复用与实例扩展,利用知识图谱对本体进行存储展示;应用知识图谱绘制构件组合“铺作”的3D模型,设计原型系统展示图谱数据,验证所建本体的应用价值。

中国风景园林理论体系建设回顾与展望

【目的】梳理20世纪80年代以来中国风景园林理论体系研究的内容及发展脉络,综合评析风景园林理论体系建设的发展水平,以期为进一步推动风景园林理论体系建设提供思路,从而促进学科理论、方法和技术的发展。【方法】从理论体系建构的不同视角或维度出发,归纳、比较和反思中国风景园林理论体系建设的途径、思路和方法,发掘看待风景园林理论与实践相关现象的不同方式。【结果】辨别和剖析了传统体系、设计体系、空间体系、知识体系、学科体系5种理论体系的异同与关联,以及对学科理论发展的影响和作用,并剖析了学科理论发展过程中存在的诸多问题。【结论】理论体系的建构能够促进对学科内在和外在关系的思考,可视为中国风景园林学科与专业理论发展的结晶,借此把握学科理论的发展动向,并以开放式的答案与认知面对当前的困境。

基于产业链的电力设备智能故障诊断方法研究

电力设备故障是导致电网无法正常运行的重要原因。针对电力设备故障诊断机制因缺乏协同管理和知识共享导致的效率低下和技术更新受限问题,提出一种基于产业链的电力设备故障诊断方法。首先基于设备故障诊断历史数据,采用BERT-BiLSTM-ATT深度神经网络构建设备故障知识图谱;然后从上下游协同、虚实互补两个方面构建电力设备故障诊断跨时空信息联动模式。该方法解决了电力设备故障根源追溯、处理措施检索难题,有助于推动电力设备的智能化、可视化故障诊断。

隐性知识深度服务体系架构及其协同机制

[研究目的]明晰需求导向下的隐性知识深度服务体系功能、要素、架构及其协同机制,提升隐性知识应用价值、改善其创新效益。[研究方法]基于深度服务理念,融合系统科学和协同学思想,从用户与环境双重视角,利用Kano模型与WBS方法剖析隐性知识服务需求、厘清其结构;在此基础上,由需求结构推演系统功能结构,基于一般系统方法论之结构方法反演隐性知识服务体系架构;进而,以WSR系统方法论分析隐性知识深度服务体系要素及其间作用机理,并基于协同学思想讨论该体系架构的协同机制。[研究结论]明确了隐性知识深度服务体系架构及其协同机制,丰富了相关理论成果,为实践域内应用系统的顶层设计提供了遵循与参考。

宜宾市0 ~ 3 岁儿童血清维生素D 水平与其监护人营养知识、态度及行为的关系

目的分析宜宾市0~3岁儿童血清维生素D水平与其监护人营养知识、态度及行为(knowledge,architecture and practice,KAP)的关系。方法采用方便抽样法,抽取2020年1月至12月于宜宾市中医医院等3所医疗机构进行儿童保健的510例0~3岁儿童为研究对象,并按年龄段分类:0~<1岁363例,1~≤3岁147例。收集喂养情况、生长发育情况、生活方式等资料。检测血清25-羟维生素D(25-hydroxy vitamin D,25-OH-D)水平。问卷调查监护人营养KAP。统计学方法采用t检验、ANOVA方差分析、χ^(2)检验、秩和检验和Pearson相关性分析。结果0~<1岁组儿童血清25-OH-D水平高于1~≤3岁组,维生素D不足和缺乏的发生率低于1~≤3岁组;体型超重/肥胖儿童的血清25-OH-D水平低于体型正常和营养不良儿童,维生素D不足和缺乏的发生率高于体型正常和营养不良组;户外活动时间≥2 h/d儿童血清25-OH-D水平高于户外活动时间<2h/d儿童,维生素D不足和缺乏的发生率低于户外活动时间<2 h/d组;维生素D补充量≥300 IU/d儿童血清25-OH-D水平高于<300 IU/d儿童,维生素D不足和缺乏的发生率低于<300 IU/d儿童,差异均有统计学意义(P值均<0.05)。监护人为父母的营养KAP评分均高于其他监护人[知识:(81.9±10.1)分与(68.8±12.5)分;态度:(88.8±7.9)分与(86.1±8.3)分;行为:(79.7±11.2)分与(72.9±12.8)分,P值均<0.05]。Pearson相关性分析结果显示,监护人营养KAP评分与血清25-OH-D水平均呈正相关(r值分别为0.801、0.647和0.836,P值均<0.05)。监护人营养KAP评分与儿童维生素D不足和缺乏有关联性,其列联系数分别为0.469、0.604和0.425。结论宜宾市0~3岁儿童血清维生素D缺乏发生率高,维生素D营养状况与监护人营养KAP评分呈正相关,提高监护人营养KAP评分有利于更好地采取合理措施防控儿童维生素D缺乏。

基于城市建成环境再生的风景园林知识体系拓展

【目的】基于新时期中国城市发展阶段的模式转型,以及风景园林学科的学科定位调整,拓展面向建成环境再生的风景园林知识体系,有助于进一步发挥风景园林规划设计实践在建成环境再生中的引领作用。【方法】在理论上,结合场域理论(field theory)和景观都市主义(landscape urbanism),指出城市建成环境再生范式的三大转变,即“场域”关系的视角转变、城市更新的路径转变、设计主体的身份转变。在实践上,依托从上海“一江一河”到全国“大江大河”的滨水空间营造、街区综合环境提升,以及城市基础设施复合化与景观化的大量实践,深化建成环境再生的“六维空间再生体系”。【结果】强化理论与实践层面对风景园林知识体系拓展的影响,包括空间营造的系统性知识、历史文脉的文化性知识、基础设施的复合性知识、场景节点的建构性知识、环境修复的生态性知识和公共空间的艺术性知识。【结论】倡导“大设计观”下多学科、多领域、多工种的设计跨界与融合,优化范式转变下的风景园林知识体系的拓展路径,推动风景园林共同体为建成环境再生创造新价值、新范式。

DLG辅助的LiDAR点云建筑物三维建模方法

现有的城市建筑三维(3D)建模方法存在数据采集成本和模型精度难以平衡的问题。提出一种基于数字线划图(DLG)和机载激光点云数据的单体建筑物三维建模方法。通过分析建筑标准总结了建筑结构的知识规则,并将这些知识规则应用于辅助DLG和点云数据重建单体建筑的三维模型,可以准确高效地提取包含楼层和详细屋顶结构信息的单体建筑模型,实现了建筑模型的高效提取。与以往研究相比,本文将建模细节层次(LOD)从LOD-2改进为LOD-2.2,能够满足城市精细化管理的需求。这为城市建筑三维模型的大规模采集提供了一条低成本、高效率的技术路线。

基于VOSviewer和CiteSpace的历史街区建筑色彩研究热点可视化分析

文章基于知识图谱分析历史街区建筑色彩的研究现状和发展趋势。检索自2006—2022年中国知网(CNKI)刊载的历史街区建筑色彩相关研究文献进行筛选。根据年发文量、作者合作网络、机构关系、热点趋势、突现词等通过文献计量法VOSviewer和CiteSpace软件绘制进行可视化分析与客观评述,从而能够直观清晰地分析历史街区建筑色彩研究现状,并判断发展趋势。分析显示,当前历史街区建筑色彩仍没有一个很好的规范图谱,在充分对其现状进行分析后,未来应加强作者与机构的合作,总结实践经验和技术方法,为今后历史街区建筑色彩研究与发展提供借鉴。

基于元模型的装备体系知识图谱构建

针对当前军事领域知识图谱描述装备体系存在的数据规范程度不高、实体不统一和实体关系不一致等问题,提出一种基于元模型的军事领域本体模型构建方法。该方法采用体系架构元模型技术框架下的概念数据模型和逻辑数据模型定义装备体系本体,从而生成装备体系知识图谱,避免不同兵种、不同业务领域装备体系本体不一致的影响。选取协同打击任务场景构建无人机集群装备体系应用本体,导入Neo4j软件生成知识图谱。结果表明,所提方法能够满足装备体系知识图谱构建需求。

风景园林理论知识体系发展溯源

【目的】探究风景园林理论知识形成的复杂脉络,能够为学科理论知识体系的构建提供新的视野。【方法】借鉴知识考古学方法,从人文与艺术学科、自然科学、社会科学三大学科领域梳理风景园林理论知识体系的发展历程,归纳演变特点及驱动因素。【结果】1)风景园林理论知识体系发展经历了从早期以人文与艺术学科知识驱动为主,逐渐走向与自然科学、社会科学的多元交叉过程。2)不同学科视野下的风景园林理论知识发展各有特点,本质上是美学观、自然观、公平观发生了改变:人文与艺术学科视野下的学科理论知识生产由地域性审美观走向批判地域性审美观;自然科学视野下的学科理论知识生产由朴素自然观走向多元生态观;社会科学视野下的学科理论知识生产由精英式公平观走向公众公平观。3)风景园林理论知识体系发展的驱动因素是由满足使用者的审美需求转向满足人在空间中生理、心理及社会全方位的功能需求。【结论】风景园林理论知识体系未来发展还需要树立自主理论知识体系的意识、梳理其逻辑基点并建立弹性和动态调整的更新机制。