基于WebGL的三维实体语义标注系统的设计与实现

发布时间：2018-08-05 13:53

【摘要】：随着多媒体技术的不断发展,网页中元素形式呈现多样化趋势。近年来,在WebGL标准的辅助下,3D模型和场景不再依赖任何第三方插件就可以流畅的渲染到浏览器中。通过标准的、跨平台的WebGL接口在浏览器中实现三维实体语义标注来替代传统的二维绘图软件标注显得十分必要。本文基于WebGL标准对三维实体语义标注系统进行设计与实现。描述了系统的需求分析、系统总体设计及数据设计、系统详细设计及系统的实现与测试等内容。首先从功能性需求与非功能性需求两个方面对系统进行分析,并确定了系统的六大功能模块。在总体设计阶段明确提出模块化、B/S结构、组件化的设计原则,确定了系统设计的总体框架,采用功能架构图对系统功能进行展示。对系统涉及到的内容进行建模分析并绘制E-R图来展示实体间的联系。详细设计时从系统的六大功能模块着手,将各模块再按照功能进行划分,对每个功能点的实现函数进行详细设计。系统实现部分围绕主要功能模块的核心代码展示及效果截图的方式展开,最后对系统进行了全面的测试并针对测试问题进行完善。通过全面的分析、详细的设计及不断的完善,最终设计出三维实体语义标注系统。本开发主要实现三维实体语义标注系统的用户管理、实体文件上传、语义标注、文件共享、文件汇总、微信接入六大功能模块。系统设计中采用Vue的组件化开发,实现了数据的绑定,极大提高了代码复用和开发效率。对于频繁请求的文件使用redis做数据缓存,减少请求次数与请求响应时间。利用H5的离线存储机制,使用户在离线状态下可继续操作,在有网络链接的情况下再将这些操作进行提交保存。为使系统更具通用性与实用性,系统完全兼容web端与手持设备端,在微信公众号下可同样使用。系统实现的语义化的标注强化了用户对实体文件的理解;在线的标注模式大大节省了文件存储空间,让用户随时随地可操作;利用语义化的标注内容可进一步优化系统内文件的检索。目前,实现的三维实体语义标注系统已经得到很好的应用,既方便了用户的学习与交流,又节省了用户本地存储空间。
[Abstract]:With the continuous development of multimedia technology, the element form of web pages presents a diversified trend. In recent years, with the help of the WebGL standard, 3D models and scenes can be rendered smoothly into browsers without relying on any third-party plug-ins. It is necessary to implement 3D entity semantic annotation in browser by standard, cross-platform WebGL interface instead of traditional two-dimensional drawing software annotation. This paper designs and implements a three-dimensional entity semantic annotation system based on WebGL standard. The requirements analysis, system design and data design, system detailed design, system implementation and test are described. First, the system is analyzed from two aspects of functional requirements and non-functional requirements, and six functional modules of the system are determined. In the stage of overall design, the modular B / S structure and the design principle of componentization are put forward, and the overall frame of the system design is determined, and the function of the system is demonstrated by the functional architecture diagram. Modeling and analyzing the contents involved in the system and drawing E-R diagram to show the relationship between entities. The detailed design starts with the six function modules of the system, and then divides the modules according to the function, and designs the realization function of each function point in detail. The realization part of the system revolves around the core code display of the main function module and the way of the effect screenshot. Finally, the system is tested comprehensively and the test problem is perfected. Through comprehensive analysis, detailed design and continuous improvement, a 3D entity semantic annotation system is designed. This development mainly realizes the user management, entity file upload, semantic annotation, file sharing, file collection and WeChat access of the 3D entity semantic annotation system. In the design of the system, the component development of Vue is used to realize the binding of data, which greatly improves the efficiency of code reuse and development. For frequently requested files, redis is used as data cache to reduce the number of requests and request response time. By using the offline storage mechanism of H5, the user can continue to operate in the offline state, and then submit these operations to save when there is a network link. In order to make the system more universal and practical, the system is fully compatible with web and handheld devices. The semantic annotation realized by the system strengthens the user's understanding of the entity file, the online annotation mode saves the file storage space greatly, and the user can operate at any time and anywhere. The retrieval of files in the system can be further optimized by using semantic annotated content. At present, the implemented 3D entity semantic annotation system has been applied very well, which not only facilitates the user's learning and communication, but also saves the user's local storage space.
【学位授予单位】：山东师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.52

【参考文献】