A highly reliable encoding and decoding communication framework based on semantic information

Increasing research has focused on semantic communication,the goal of which is to convey accurately the meaning instead of transmitting symbols from the sender to the receiver.In this paper,we design a novel encoding and decoding semantic communication framework,which adopts the semantic information and the contextual correlations between items to optimize the performance of a communication system over various channels.On the sender side,the average semantic loss caused by the wrong detection is defined,and a semantic source encoding strategy is developed to minimize the average semantic loss.To further improve communication reliability,a decoding strategy that utilizes the semantic and the context information to recover messages is proposed in the receiver.Extensive simulation results validate the superior performance of our strategies over state-of-the-art semantic coding and decoding policies on different communication channels.

Adaptive Boundary and Semantic Composite Segmentation Method for Individual Objects in Aerial Images

There are two types of methods for image segmentation.One is traditional image processing methods,which are sensitive to details and boundaries,yet fail to recognize semantic information.The other is deep learning methods,which can locate and identify different objects,but boundary identifications are not accurate enough.Both of them cannot generate entire segmentation information.In order to obtain accurate edge detection and semantic information,an Adaptive Boundary and Semantic Composite Segmentation method(ABSCS)is proposed.This method can precisely semantic segment individual objects in large-size aerial images with limited GPU performances.It includes adaptively dividing and modifying the aerial images with the proposed principles and methods,using the deep learning method to semantic segment and preprocess the small divided pieces,using three traditional methods to segment and preprocess original-size aerial images,adaptively selecting traditional results tomodify the boundaries of individual objects in deep learning results,and combining the results of different objects.Individual object semantic segmentation experiments are conducted by using the AeroScapes dataset,and their results are analyzed qualitatively and quantitatively.The experimental results demonstrate that the proposed method can achieve more promising object boundaries than the original deep learning method.This work also demonstrates the advantages of the proposed method in applications of point cloud semantic segmentation and image inpainting.

A New Behavior-Assisted Semantic Recognition Method for Smart Home

In the recent Smart Home(SH)research work,intelligent service recommendation technique based on behavior recognition,it has been extensively preferred by researchers.However,most current research uses the Semantic recognition to construct the user’s basic behavior model.This method is usually restricted by environmental factors,the way these models are built makes it impossible for them to dynamically match the services that might be provided in the user environment.To solve this problem,this paper proposes a Semantic behavior assistance(Semantic behavior assistance,SBA).By joining the semantic model on the intelligent gateway,building an SA model,in this way,a logical Internet networks for smart home is established.At the same time,a behavior assistant method based on SBA model is proposed,among them,the user environment-related entities,sensors,devices,and user-related knowledge models exist in the logical interconnection network of the SH system through the semantic model.In this paper,the data simulation experiment is carried out on the method.The experimental results show that the SBA model is better than the knowledge-based pre-defined model.

Semantic and Communicative Translation:On the English Translation Methods of Martial Arts Moves in Legends of the Condor Heroes

The martial arts novel is not only a significant kind of Chinese popular fictions but also a novel category unique to Chi⁃nese culture.Legends of the Condor Heroes is the representative one which was once adapted into nationally popular TV series,boasting a widespread domestic mass base.Therefore,the book may become a key to the door of foreign readers’interest in Chi⁃nese martial arts culture.This thesis takes the latest English version of Legends of the Condor Heroes as the research object,focus⁃ing on six specific translation methods from the perspectives of semantic translation and communicative translation.The six transla⁃tion methods discussed in the thesis include:(1)literal translation;(2)combination of transliteration and literal translation;(3)liter⁃al translation with necessary notes;(4)free translation;(5)combination of transliteration and free translation;(6)creative transla⁃tion.

Building Extraction from LIDAR Based Semantic Analysis

Extraction of buildings from LIDAR data has been an active research field in recent years. A scheme for building detection and reconstruction from LIDAR data is presented with an object-oriented method which is based on the buildings’ semantic rules. Two key steps are discussed: how to group the discrete LIDAR points into single objects and how to establish the buildings’ semantic rules. In the end, the buildings are reconstructed in 3D form and three common parametric building models (flat, gabled, hipped) are implemented.

Semantic Complex Event Detection System of Express Delivery Business with Data Support From Multidimensional Space

The express delivery induslry of China is relatively backward in the automation degree of critical business processes. The basic reason is that the business-related supporting data, which is scattered in the multidimensional space, is difficult to utilize and process. This paper proposes an automatic data acquisition fi-amework to resolve such difficulty, which synthetically utilize intelligent inemet of things （IoT）, semantic web and complext event processing （CEP） technology. We also implement a SCEP prototype system with the capability of real-time detecting complex business events on the goods sorting line, which adopts a detection method consisting of four stages. The simulation results show that the system has good performance and feasible enough to deal with the complex business which need data support fTom multidimensional space.

Heuristics based semantic annotation of biodiversity documents in Chinese

Purpose: To design an efficient high-performance algorithm for semantic annotation of biodiversity documents in Chinese.Design/methodology/approach: Data set consists of 1,000 randomly selected documents from Flora of China. Comparative evaluation of the proposed approach with the Na ve Bayes algorithm have been developed before for the same purpose.Findings: Experimental results show that the heuristics based algorithm outperformed the Na ve Bayes algorithm. The use of leading words helped improving the annotation performance while prioritizing rule application based on their weights had no significant impact on algorithm performance.Research limitations: The ICTCLAS was used to identify word boundaries off-shelf without optimatization for biodiversity domain. This may have not made the best use of the tool.Practical implications & Originality/value: The performance of heuristics based approach,enhanced by leading words analysis, reached an F value of 0.9216, which is sufficiently accurate for practical use.

改进DeepLabV3+算法提取无作物田垄导航线

机器视觉导航是智慧农业的重要部分,无作物田垄的导航线检测是旱地移栽导航的关键。针对无作物田垄颜色信息相近、纹理差距小,传统图像处理方法适用性差、准确率低,语义分割算法检测速度慢、实时性差的问题,该研究提出一种基于改进DeepLabV3+的田垄分割模型。首先对传统DeepLabV3+网络进行轻量化设计,用MobileNetV2网络代替主干网络Xception,以提高算法的检测速度和实时性;接着引入CBAM(convolutional block attention module,CBAM)注意力机制,使模型能够更好地处理垄面边界信息;然后利用垄面边界信息获得导航特征点,对于断垄情况,导航特征点会出现偏差,因此利用四分位数对导航特征点异常值进行筛选,并采用最小二乘法进行导航线拟合。模型评估结果显示,改进后模型的平均像素精确度和平均交并比分别为96.27%和93.18%,平均检测帧率为84.21帧/s,优于PSPNet、U-Net、HRNet、Segformer以及DeepLabV3+网络。在不同田垄环境下,最大角度误差为1.2°,最大像素误差为9,能够有效从不同场景中获取导航线。研究结果可为农业机器人的无作物田垄导航提供参考。

Exploring Taiwan’s China Landscape Painting Aesthetic Preferences Through Evaluation Grid Method and the Continuous Fuzzy Kano Model

The primary objective of this study is to apply the Evaluation Grid Method(EGM)and the continuous fuzzy Kano quality model to explore the cognitive preferences of Taiwan China residents regarding the beauty of Taiwan’s China landscape paintings.The aim is to contribute to the development of social and cultural art and promote the widespread appeal of art products.Through a literature review,consultations with aesthetic experts,and the application of Miryoku Engineering’s EGM,this paper consolidates the factors that contribute to the attractiveness of painting art products among Taiwan China residents,taking into account various aesthetic qualities.Simultaneously,the paper introduces the use of the triangular fuzzy golden ratio scale semantics,specifically the equal-ratio aesthetic scale semantics,as a replacement for the traditional subjective consciousness model.Departing from the traditional discrete Kano model that employs the mode as the standard for evaluating quality,this study applies triangular fuzzy numbers to the continuous Kano quality model to analyze the diverse preferences and evaluation standards of the public.The hope is that this research methodology will not only deepen Taiwan China residents’understanding and aesthetic literacy of painting art but also serve as a reference for the popularization of art products.

基于语义分析法的城市街景色彩感知评价与优化研究

为探究使用者对哈尔滨市中央大街现状色彩的感知评价情况,该文运用语义分析法这一采用“语言”方式对公众进行心理测评的定量分析工具,结合街景图像、色卡软件、数理分析软件作为记录和分析色彩样本的基础工具对其开展问卷调查与研究。研究结果表明,受访者对中央大街街区的街景色彩总体持正面评价,对色彩视觉感知体验更好,空间感知体验最差,对经过系统色彩设计、风格鲜明的街道评价更高。依据以上结果,从总体色彩构成、一般街道品质提升和空间感知强化等方面提出了街景色彩优化策略。

基于墙角族语义尺寸链的绑架定位研究

针对目前原始自适应蒙特卡洛定位(Adaptive Monte Carlo Localization,AMCL)在相似环境下绑架检测容易出错且重定位极易失败等问题,提出基于墙角族语义尺寸链的改进AMCL算法.融合机器人多传感器信息和Gmapping算法构建二维栅格地图,基于Yolov5获取室内环境的目标检测框和类别信息,结合GrabCut算法和贝叶斯方法构建增量式语义映射地图;通过墙角的凸、凹和墙角相对于机器人的方位角对墙角进行分类,充分发掘语义映射地图中各墙角之间、墙角与室内物体之间的类别和位置关系,构建墙角族语义尺寸链和相应检索表;在定位过程中,基于墙角族语义尺寸链进行全局预定位,提出绑架检测机制进行绑架检测,在检测到绑架事件发生后,基于改进AMCL算法实现定位自恢复.最后,通过真实环境下的绑架实验验证了本文方法的有效性,实验表明,所提方法的全局定位准确率、全局定位速率、绑架检测准确率和绑架后定位准确率在相似环境下分别提升了42%、214%、88%和72%;在非相似环境下分别提升了44%、152%、12%和92%;在长走廊环境下分别提升了36%、426%、26%和68%.

融合模糊层次分析法和特征语义法的交流往复锯形态设计研究

通过对不同品牌交流往复锯的形态分析,明确了用户对交流往复锯的核心需求,包括产品品质、设计美学和功能优化等方面。通过特征语义法提取了形态特征语义词,利用层次分析法构建了需求指标体系,计算了各指标的权重,以确定设计中应优先考虑的因素,并将其应用于设计方案的生成。设计方案以专业级工具中的“绅士”为定位,强调平稳、高效以及自然与机械的和谐共生。采用模糊综合评价法对设计方案进行了评价,验证其有效性和可行性。评价结果显示,设计方案达到了较为满意的状态,为后续的设计改进提供了明确方向。

基于信息检索的源代码自动命名

源代码自动命名是指为给定代码的方法体命名一个反映代码功能的有意义的名称,可以使代码易读易懂,提高软件开发效率。传统自动命名方法仅使用代码的词法或者语法等单一信息,基于深度学习的自动命名方法通常忽略了语料库中的相似代码,影响命名准确率。针对上述问题,提出一种基于信息检索的源代码自动命名方法。首先,利用预训练模型和BERT-whitening方法提取输入代码和语料库中代码的有效特征,使用欧氏距离计算两者之间的语义相似度。其次,在语料库代码中选择与输入代码语义相似度较高的代码组成候选库,利用Jaccard系数和最长公共子序列分别计算输入代码与候选库代码的词法和语法相似度。最后,使用加权和来匹配候选库中与输入代码最相似的代码片段,复用该代码片段的方法名称作为输入代码的方法名称。实验结果表明,在公开的Java-small数据集上,与基于向量空间模型(VSM)和基于深度学习模型Code2Vec的自动命名方法相比,该方法的F1值分别提升了6.93和1.22个百分点,具有较优的预测性能。

基于AHP--SD法的圭塘河滨水景观评价研究

以长沙市圭塘河的滨水景观作为研究对象,从景观生态、服务管理和社会文化3个方面来对滨水景观作综合评价。运用层次分析法(AHP)确定各指标的权重,再利用语义差异法(SD)整体分析并发放调查问卷获得使用者们评价的基本数据。结果表明,准则层分值由高到低依次是景观生态、服务给管理和社会文化;在18个指标层因子中,主要影响滨水景观的重要因子有河流水质、滨水植被、驳岸设计和植物保护,该圭塘河滨水景观最终评分67.68分,评价为一般满意。对圭塘河当地的特色文化的表现程度和公共设施的建设有待加强,滨水步道应优化建设。

基于实体识别的医嘱复合词标准化方法的应用研究

目的:提高医嘱复合词数据清洗和数据标准化的效率。方法:收集2015年―2021年在解放军总医院急救患者的住院医嘱数据,共计22809人,27096例次。以知识库为基础,利用分级词典将医嘱复合词拆解为不同层级的实体,基于语义重组算法对实体进行笛卡尔积重组,重组结果做标准化处理,最终辅以临床医学专家对标准化处理结果人工校验。结果:针对医嘱复合词,如检查和手术类医嘱的标准化效果良好。结论:本方法提高了医嘱复合词标准化的准确率,节省了数据治理的时间。

基于实例分割与光流的动态环境SLAM

针对传统语义实时定位与建图(Semantic real-time Localization and Mappling,SLAM)算法在动态环境下剔除特征点过多、造成定位精度降低的问题,提出一种基于实例分割与光流的视觉语义SLAM算法。算法使用Mask R-CNN网络对图像中的潜在动态物体进行实例级别的分割,同时在光流线程中对动态物体进行识别并剔除,随后使用剩余的静态光流点与静态特征点联合优化定位,实现语义信息与光流信息的充分融合利用。使用公开数据集测试和地面无人平台实验对所提方法进行验证。实验结果表明,在TUM数据集下,新方法的定位均值误差相比ORB-SLAM2平均提高75%,相比Dyna-SLAM平均提高8.5%。

基于语义不一致性的网络暴力舆情预警方法

[研究目的]现有网络暴力舆情预警手段对专家知识和先验事件信息存在较大依赖。该文提出一种基于语义不一致性的网络暴力舆情预警方法,以实现对现实情境中网络暴力舆情的有效预警。[研究方法]该文采用网络暴力舆情预警的语义不一致性指数计算方法,并结合时序异常检测方法对语义不一致性异常波动进行监测并预警。收集2022年7月至9月间的微博数据进行模拟预警,验证模型预警能力。[研究结论]实验中平均预警准确率为76.57%,预警覆盖率为60.92%。经验证可实现真实世界条件下至多提前36小时发出预警。实验揭示了网络暴力舆情事件在发展阶段可对整体内容的主题状态产生显著可监测的影响,基于该特性能够实现或增强舆情预警感知能力。

基于语义差分法和模糊综合评价法的校园吉祥物创新设计方法研究

研究目的在于探究满足用户感性需求的校园吉祥物设计,提升满意度及认同感,丰富校园文化建设。首先,通过调查分析等方式确定现有校园吉祥物的典型样本及关键感性词汇;其次,运用语义差分法量化用户的感性认知,结合因子分析及主成分分析提取主要词汇,并以华南理工大学的铭诚书院为对象开展设计实证;最后,结合模糊综合评价法验证实践可行性及满意度。通过提取用户对校园吉祥物的感性认知,可引导设计者更全面了解用户感性需求,设计更易于被接受、贴近预期满意度的校园吉祥物,增强师生认同感及校园文化影响力,为校园吉祥物提供创新设计思路。

基于语义分割和融合算法的草坪杂草定位

【目的】探究基于多算法融合的草坪杂草精准定位算法,为草坪杂草的自动识别和清除机器人的除草作业提供技术支撑与理论依据。【方法】提出了一种基于语义分割的多算法融合的草坪定位算法。首先,通过PSPNet网络分割草坪和非草坪轮廓。其次,针对分割出来的非草坪轮廓提取感兴趣的区域,去除非杂草轮廓,保留杂草轮廓。然后,利用Zhang-Suen细化算法提取杂草轮廓骨架线,并获取骨架交叉点数量和坐标位置。最后,利用融合算法依据交叉点数量选择不同的定位策略,实现杂草根部的精准定位。【结果】融合算法定位的杂草坐标与真实杂草根部中心坐标的均方根误差为83.17像素,相比平均质心法减少了14%,相比最小外接圆减少了22%。换算到实际场景之下,融合算法定位的杂草坐标与真实杂草根部中心坐标的均方根误差为12.48 mm,误差在可接受的误差范围内。融合算法提高了杂草根部中心的定位精度,降低了杂草定位的误差。【结论】基于语义分割和融合算法的草坪杂草定位方法提高了杂草根部中心的定位精度,降低了单一方法的定位误差,可以为草坪杂草自动识别和除草作业提供技术支持。

科技情报中的多源数据融合方法研究综述

[研究目的]多源数据融合可通过交叉引证降低信息不一致和语义模糊等不确定因素,面对大数据和人工智能的普及导致的海量多源异构数据急剧增加,如何对其进行有效融合成为了科技情报领域研究的重难点之一。[研究方法]首先根据现有定义对多源数据的概念特征进行阐释;其次按照对文本中数据融合的深入程度,对多源数据融合方法进行系统梳理;最后分析了多源数据融合在不同科技情报场景中的应用和发挥的作用。[研究结论]从概念来看,多源数据融合具有内源性和外源性两种指向。从融合方法来看,现有常用方法主要包括三种:数据源的融合、结构关系融合以及语义融合,融合方法在实现从物理融合向化学融合转变的同时也在科技发展趋势研判、学术评价、需求探测等场景下得到广泛应用。未来研究需要通过实践操作细化异构加权的可行方法,以应用为导向加强对于融合方法的总结和归纳,探索形成可供复用的模型框架。