Improving the Effectiveness of Image Classification Structural Methods by Compressing the Description According to the Information Content Criterion

The research aims to improve the performance of image recognition methods based on a description in the form of a set of keypoint descriptors.The main focus is on increasing the speed of establishing the relevance of object and etalon descriptions while maintaining the required level of classification efficiency.The class to be recognized is represented by an infinite set of images obtained from the etalon by applying arbitrary geometric transformations.It is proposed to reduce the descriptions for the etalon database by selecting the most significant descriptor components according to the information content criterion.The informativeness of an etalon descriptor is estimated by the difference of the closest distances to its own and other descriptions.The developed method determines the relevance of the full description of the recognized object with the reduced description of the etalons.Several practical models of the classifier with different options for establishing the correspondence between object descriptors and etalons are considered.The results of the experimental modeling of the proposed methods for a database including images of museum jewelry are presented.The test sample is formed as a set of images from the etalon database and out of the database with the application of geometric transformations of scale and rotation in the field of view.The practical problems of determining the threshold for the number of votes,based on which a classification decision is made,have been researched.Modeling has revealed the practical possibility of tenfold reducing descriptions with full preservation of classification accuracy.Reducing the descriptions by twenty times in the experiment leads to slightly decreased accuracy.The speed of the analysis increases in proportion to the degree of reduction.The use of reduction by the informativeness criterion confirmed the possibility of obtaining the most significant subset of features for classification,which guarantees a decent level of accuracy.

Adansonia digitata L.(baobab):a review of traditional information and taxonomic description

Adansonia digitata L.(Malvaceae) is commonly known as baobab tree native to Africa.Baobab is a multi-purpose tree which offers protection and provides food,clothing and medicine as well as raw material for many useful items.The fruit pulp,seeds,leaves,flowers.roots,and bark of baobab are edible and they have been studied by scientists for their useful properties.The fruit pulp have very high vitamin C.calcium,phosphorus,carbohydrates,fibers,potassium,proteins and lipids content,which can be used in seasoning as an appetizer and also make juices.Seeds contain appreciable quantities of phosphorus,magnesium,zinc,sodium,iron,manganese,whereas they have high levels of lysine,thiamine,calcium and iron.Baobab has numerous biological properties including antimicrobial,anti-malarial,diarrhoea,anaemia,asthma,antiviral,anti-oxidant and anti-inflammatory activities amongst others.Phytochemical investigation revealed the presence of flavonoids.phytosterols.amino acids,fatty acids,vitamins and minerals.The review summarizes the information on various aspects of traditional information,taxonomic description,medicinal properties and importantly nutritional value.

Exploring the Characteristics of Data Breaches: A Descriptive Analytic Study

Data breaches have massive consequences for companies, affecting them financially and undermining their reputation, which poses significant challenges to online security and the long-term viability of businesses. This study analyzes trends in data breaches in the United States, examining the frequency, causes, and magnitude of breaches across various industries. We document that data breaches are increasing, with hacking emerging as the leading cause. Our descriptive analyses explore factors influencing breaches, including security vulnerabilities, human error, and malicious attacks. The findings provide policymakers and businesses with actionable insights to bolster data security through proactive audits, patching, encryption, and response planning. By better understanding breach patterns and risk factors, organizations can take targeted steps to enhance protections and mitigate the potential damage of future incidents.

Empirical Evidence for a Descriptive Model of Principles of Information Security Course

The purpose of this study is to examine the nature and content of the rapidly evolving undergraduate Principles of Information/Cybersecurity course which has been attracting an ever-growing attention in the computing discipline, for the past decade. More specifically, it is to provide an impetus for the design of standardized principles of Information/Cybersecurity course. To achieve this, a survey of colleges and universities that offer the course was conducted. Several schools of engineering and business, in universities and colleges across several countries were surveyed to generate necessary data. Effort was made to direct the questionnaire only to Computer Information System (CIS), Computer Science (CS), Management Information System (MIS), Information System (IS) and other computer-related departments. The study instrument consisted of two main parts: one part addressed the institutional demographic information, while the other focused on the relevant elements of the course. There are sixty-two (62) questionnaire items covering areas such as demographics, perception of the course, course content and coverage, teaching preferences, method of delivery and course technology deployed, assigned textbooks and associated resources, learner support, course assessments, as well as the licensure-based certifications. Several themes emerged from the data analysis: (a) the principles course is an integral part of most cybersecurity programs;(b) majority of the courses examined, stress both strong technical and hands-on skills;(c) encourage vendor-neutral certifications as a course exit characteristic;and (d) an end-of-course class project, remains a standard requirement for successful course completion. Overall, the study makes it clear that cybersecurity is a multilateral discipline, and refuses to be confined by context and content. It is envisaged that the results of this study would turn out to be instructive for all practical purposes. We expect it to be one of the most definitive descriptive models of such a cardinal course, and help to guide and actually, shape the decisions of universities and academic programs focusing on information/cyber security in the updating and upgrading their curricula, most especially, the foundational principles course in light of new findings that are herein articulated.

Chaining Distributed Geographic Information Web Services Based on Ontology

To solve the problem of chaining distributed geographic information Web services （GI Web services）, this paper provides an ontology-based method. With this method, semantic service description can be achieved by semantic annotation of the elements in a Web service description language（WSDL） document with concepts of geographic ontology, and then a common under-standing about service semantics between customers and providers of Web services is built. Based on the decomposition and formalization of customer requirements, the discovery, composition and execution of GI Web services are explained in detail, and then a chaining of GI Web services is built and used to achieve the customer＇s requirement. Finally, an example based on Web ontology language for service （OWL-S） is provided for testing the feasibility of this method.

Multi-Level Feature Aggregation-Based Joint Keypoint Detection and Description

Image keypoint detection and description is a popular method to find pixel-level connections between images,which is a basic and critical step in many computer vision tasks.The existing methods are far from optimal in terms of keypoint positioning accuracy and generation of robust and discriminative descriptors.This paper proposes a new end-to-end selfsupervised training deep learning network.The network uses a backbone feature encoder to extract multi-level feature maps,then performs joint image keypoint detection and description in a forward pass.On the one hand,in order to enhance the localization accuracy of keypoints and restore the local shape structure,the detector detects keypoints on feature maps of the same resolution as the original image.On the other hand,in order to enhance the ability to percept local shape details,the network utilizes multi-level features to generate robust feature descriptors with rich local shape information.A detailed comparison with traditional feature-based methods Scale Invariant Feature Transform(SIFT),Speeded Up Robust Features(SURF)and deep learning methods on HPatches proves the effectiveness and robustness of the method proposed in this paper.

Definition,measuring methods and status of carbon information disclosure in Chinese enterprises

From the perspective of carbon disclosure goal,the paper defined the detail types of contents about carbon information disclosure,then constructed the carbon information disclosure index,by using the indices to 2011 annual report of A-share listed companies in Shanghai and Shenzhen Security Exchange,we found that Chinese enterprises' carbon disclosure scattered,industry structure quite different characteristics,quantity information and quality information asymmetry.

基于Transformer网络多模态融合的密集视频描述方法

针对目前的密集视频描述模型大多使用两阶段的方法存在效率较低、忽略音频及语义信息,描述结果不全面的问题。提出了一种基于Transformer网络多模态和语义信息融合的密集视频描述方法。提取自适应R(2+1)D网络提取视觉特征,设计了语义探测器生成语义信息,加入音频特征进行补充,建立了多尺度可变形注意力模块,应用并行的预测头,加快模型收敛速度,提高模型精度。实验结果表明:模型在2个基准数据集上性能均有很好的表现,评价指标BLEU4上达到了2.17。

基于Web 2.0的软件工程实践教学平台设计

软件工程实践教学平台在运行过程中,存在没有全方位整合教学资源的问题。为此,提出基于Web 2.0的软件工程实践教学平台设计研究。将恩智浦iMX8MM开发板DB11作为平台硬件装置,构建Web 2.0技术为支撑的软件工程实践教学平台构架,动态整合工程实践教学资源,并引入单参数项目信息函数作为软件工程实践教学资源的属性进行划分。在测试结果中,进程/线程交换率不仅能够结合并行请求规模实现适应性调整,而且调整的幅度较大。

基于潜在特征增强网络的视频描述生成方法

视频描述生成旨在用自然语言描述视频中的物体及其相互作用。现有方法未充分利用视频中的时空语义信息,限制了模型生成准确描述语句的能力。为此,提出一种用于视频描述生成的潜在特征增强网络(LFAN)模型。利用不同的特征提取器提取外观特征、运动特征和目标特征,将对象级的目标特征分别和帧级的外观特征与运动特征融合,同时对融合后的不同特征进行增强,在生成描述前利用图神经网络和长短时记忆网络推理对象之间的时空关系,从而得到具有时空信息和语义信息的潜在特征,同时使用长短时记忆网络和门控循环单元的解码器生成视频的描述语句。该网络模型能够准确地学习到对象特征,进而引导生成更准确的词汇及与对象之间的关系。在MSVD和MSR-VTT数据集上的实验结果表明,LFAN模型可以显著提高生成描述语句的准确性,并与视频中的内容呈现出更好的语义一致性,在MSVD数据集上的BLEU@4和ROUGE-L分数分别为57.0和74.1,在MSRVTT数据集上分别为43.8和62.1。

信息空间理论下的文献遗产数字化传播研究

传统的文献遗产数字化传播缺乏相关的理论支持和数字化传播路径。为此,本文基于信息空间理论提出文献遗产数字化传播的方法框架,分析编码、抽象和扩散三维空间和采邑、宗族、官僚、市场四个典型区域的文献遗产数字化传播特性,并对文献遗产数字化传播进行规划指导。本文探索出“文献遗产数字化—文献遗产资源描述和语义组织—文献遗产资源有效传播”的数字化传播链条,实现从采邑区、宗族区、官僚区到市场区的数字化传播动态循环与创新过程。

大数据环境下病案信息协同管理模型构建

针对大数据环境下病案信息管理呈现的新特征,文中采用计算机系统结构方法,以病案信息为构建实体,在遵循集中-分布式构建规范、三部分协同实现构建原理的基础之上,按照不同数据类型内部结构组织方式不同的原则,根据客户需求反向推动机制工作流程,从信息交流、组织结构、通信传递、功能实现、管理协同五个维度抽象构建了病案信息协同管理模型。以期为病案信息管理、模型构建相关研究提供理论借鉴和参考。

煤矿GIS一张图快速构建平台研究

目前,煤矿使用的工程专题地图基本都是CAD制图,高效提取CAD图件中的数据并快速组织成地理信息系统(GIS)服务,进而支持矿井空间对象创建和业务属性扩展,集成安全生产实时数据,是构建煤矿GIS一张图的关键。基于ArcGIS平台将CAD图件转为GIS服务的方法实现过程较为繁琐,且ArcGIS平台成本较高,无法较好地跨平台运行。针对该问题,设计了一种煤矿GIS一张图快速构建平台。该平台包括CAD图件管理、地图服务发布、专题地图管理3大功能模块:CAD图件管理模块用于图件基础信息管理和状态跟踪,地图服务发布模块用于地图打包发布和在线预览,专题地图管理模块用于地图服务管理、矿井对象创建及属性扩展。基于开放设计联盟(ODA)的Teigha for Java SDK实现CAD图件全要素精确识别与快速准确提取;通过构建基于GIS数据特征的煤矿CAD图件数据分层描述模型,实现了CAD图件全要素数据快速存储;按照面向对象设计思路,开发了Spring Cloud框架下的Web端煤矿CAD图件数据集存储接口及专题地图服务发布平台,实现了煤矿GIS一张图快速构建。以某煤矿采掘工程平面图为例,分别采用传统方法和快速构建平台进行煤矿GIS一张图的构建,对比结果表明,快速构建平台可大幅提高煤矿GIS一张图的构建效率,为煤矿智能化建设提供时空数字底座。

三维建模软件和AutoCAD交互作用在画法几何教学中的应用探索

针对画法几何课程截交线、相贯线、组合体的三视图表达等内容教学过程中,学生对相关几何元素的空间特征和平面投影理解困难,且传统的讲解方法手工作图烦琐、知识输出量少等情况。为了提高画法几何课程的实践性及应用性,基于信息化教学手段,提出一种三维建模软件和AutoCAD交互作用的运用方法。首先基于三维建模软件对三维模型进行建模,通过立体展示增加学生对相关几何元素的空间想象和理解,同时通过AutoCAD作图演示代替传统的手工作图,提高了作图演示的效率。最后结合课堂实际教学环节,针对三维建模和AutoCAD交互作用在画法几何教学中的应用提出相关注意事项。

A description logic based approach to formalizing the information bearing capability of paths in ER schemata

Purpose–The purpose of this paper is to show how description logics(DLs)can be applied to formalizing the information bearing capability(IBC)of paths in entity-relationship(ER)schemata.Design/methodology/approach–The approach follows and extends the idea presented in Xu and Feng(2004),which applies DLs to classifying paths in an ER schema.To verify whether the information content of a data construct(e.g.a path)covers a semantic relation(which formulates a piece of information requirement),the principle of IBC under the source-bearer-receiver framework is presented.It is observed that the IBC principle can be formalized by constructing DL expressions and examining constructors(e.g.quantifiers).Findings–Description logic can be used as a tool to describe the meanings represented by paths in an ER schema and formalize their IBC.The criteria for identifying data construct distinguishability are also discovered by examining quantifiers in DL expressions of paths of an ER schema.Originality/value–This paper focuses on classifying paths in data schemas and verifying their formalized IBC by using DLs and the IBC principle.It is a new point of view for evaluation of data representation,which looks at the information borne by data but not data dependencies.

基于OPC UA的数字孪生矿井提升系统信息模型研究

针对矿井提升系统中子系统种类多、设备类型繁杂、数据多源异构,具有语义不同且碎片化的特点,提出了一种基于OPC UA的数字孪生矿井提升系统信息模型构建方法。首先建立了基于OPC UA的数字孪生数据集成体系架构,根据OPC UA建模规则构建矿井提升机系统信息模型,然后通过建模工具UaModeler生成XML描述文件,通过搭建OPC UA服务器将信息模型结构数据与属性集映射在OPC UA地址空间。最后利用OPC UA客户端工具UaExpert连接服务器对信息模型进行测试。结果表明:基于OPC UA的数字孪生矿井提升系统信息模型可以实现多源异构数据的虚实映射,获取信息模型结构与对象数据,实时在数字孪生可视化平台上显示,验证了信息模型的有效性,为矿井提升数字孪生系统提供数据交互服务,实现矿井提升系统的状态监控,故障预警等。

融合实体描述与路径信息的知识图谱表示学习模型

知识图谱表示学习方法是将知识图谱中的实体和关系通过特定规则表示成一个多维向量的过程。现有表示学习方法多用于解决单跳知识图谱问答任务,其多跳推理能力无法满足实际需求,为提升多跳推理能力,提出一种融合实体描述与路径信息的知识图谱表示学习模型。首先通过预训练语言模型RoBERTa得到融合实体描述的实体、关系表示学习向量;其次利用OPTransE将知识图谱转化成融入有序关系路径信息的向量。最后构建总能量函数,将针对实体描述和路径信息的向量进行融合。通过实验分析与对比该模型在链路预测任务上与主流知识图谱表示学习模型的性能,验证了该模型的可行性与有效性。

ISBD(统一版)视角下《中国文献编目规则》(第2版)和《中文图书著录规则》的比较和评述

以图书为例,在参照ISBD(统一版)的基础上,从著录信息源和著录项目两个方面比较了《中国文献编目规则》(第2版)和CALIS《中文图书著录规则》著录方式存在的不同之处,建议这两种规则应互相取长补短。

面向战略性新兴产业政策制定的情报感知研究

以“情报感知”理念指导战略性新兴产业情报工作,能够帮助决策者应对复杂决策环境和科学把握产业政策窗口。本文在战略性新兴产业政策制定关切下,分析了情报用户需求、情报对象内容、情报任务组织三类情报感知要素,以情报感知线索发现、情报感知技术应用、情报感知结果刻画作为实施要件,针对战略性新兴产业决策中的信息不完备问题,提出了基于情报感知的解决方案,为开展基于情报感知理念的科技情报工作提供了参考示范。

基于语义分割的全卷积图像描述模型

为快速生成准确描述图片内容的语句,提出语义分割和卷积神经网络(convolutional neural network,CNN)相结合的图像描述方法。将图像分类模型和语义分割模型结合为编码器,增强对图像语义信息的利用,采用CNN代替长短时记忆网络(long short term memory,LSTM)作为解码器生成完整描述性语句。通过在MSCOCO数据集上与5种主流算法的对比实验可知,以CNN作为解码器能够大幅提高解码速度,语义信息的增强能够有效提高实验精度,验证了该方法的有效性和可行性。