当前位置:主页 > 文艺论文 > 广告艺术论文 >

广播电视广告监测系统的设计与实现

发布时间:2018-07-15 16:30
【摘要】:广播电视播放过程中包含着大量的广告,这些广告有可能出现漏播,少播或违规等情况,因此工商管理部门和广告商就需要对广告的播出情况进行监控。随着多媒体相关技术的发展,使得利用该技术进行广告检索成为可能。当前,广播电视节目中的广告检索技术成为研究的热点,具有重要的研究意义和应用价值。本文调研了当前的广告检索技术,把音频语谱图像识别技术应用于广告检索,并且在此基础上设计实现了B/S (Browser/Server)架构的广播电视广告监测系统。本论文主要完成以下工作: 一、对目前广告检索中使用的多媒体检索技术进行了调研,引出了利用音频语谱图像识别技术进行广告检索的想法,然后实现了一种基于音频语谱图像识别的广告检索方法。该方法把每个广告片段的音频转换成语谱图像,并从中提取可区分的、适合索引的特征。检索时,通过特征对齐,采用基于EM(期望最大化算法)的迭代方法识别最匹配的广告。实验结果表明,该方法可以快速准确地从广播电视中检索到广告,而且具有很好的鲁棒性。 二、设计了基于SSH (Spring、Struts、Hibernate)架构+YUI (Yahoo! User Interface Library)框架+FreeMarker模板的广播电视广告监测系统。在设计时进行了充分的需求分析,包括系统功能需求和非功能需求,给出了该系统的总体框架设计、接口设计、人工处理设计等。 三、实现了广播电视广告监测系统,给出了分层实现过程、部分源代码以及界面截图等。实现后的该系统具有界面美观、操作简单、稳定可靠、易扩展、实施维护方便等优点。 本文内容来自中国科学院自动化研究所研发的广播电视内容综合监测系统,该系统是用来对广播电视的播放内容进行监测,其中就包含了广告监测。本人在中科院自动化研究所实习期间负责广告监测模块的设计和研发。该系统目前已经投入使用,并且取得了很好的效果。
[Abstract]:Radio and television broadcast process contains a large number of advertisements, these advertisements may appear missing, less broadcast or illegal situation, so the business administration departments and advertisers need to monitor the broadcast of advertisements. With the development of multimedia related technology, it is possible to use this technology for advertising retrieval. At present, advertising retrieval technology in radio and television programs has become a hot spot, which has important research significance and application value. This paper investigates the current advertising retrieval technology, applies the audio spectrum image recognition technology to the advertising retrieval, and designs and implements a B / S (browser / Server) radio and television advertising monitoring system. The main work of this paper is as follows: first, the multimedia retrieval technology used in advertising retrieval is investigated, and the idea of using audio spectral image recognition technology to carry out advertising retrieval is introduced. Then an advertisement retrieval method based on audio spectrum image recognition is implemented. The method converts the audio frequency of each advertisement fragment into the idiom spectrum image and extracts distinguishing and indexed features. In retrieval, feature alignment is used to identify the most matched ads using an iterative method based on EM (expectation maximization algorithm). Experimental results show that this method can retrieve advertisements from radio and television quickly and accurately, and has good robustness. Secondly, design Yui (Yahoo!) based on SSH (Spring Struts hibernate) architecture! User Interface Library) Framework FreeMarker template radio and television advertising monitoring system. In the design, the requirement analysis is carried out, including the functional and non-functional requirements of the system. The overall framework design, interface design, manual processing design and so on are given. Third, the broadcast and television advertisement monitoring system is implemented, and the hierarchical implementation process, part of the source code and interface screenshots are given. The system has the advantages of beautiful interface, simple operation, stable and reliable, easy to expand and easy to carry out maintenance. This paper comes from the integrated monitoring system of radio and television content developed by the Institute of Automation of the Chinese Academy of Sciences, which is used to monitor the broadcast content of radio and television, including advertising monitoring. I am responsible for the design and development of advertising monitoring module during the internship in the Institute of Automation of the Chinese Academy of Sciences. The system has been put into use and has achieved good results.
【学位授予单位】:北京交通大学
【学位级别】:硕士
【学位授予年份】:2011
【分类号】:TN948.1

【参考文献】

相关期刊论文 前5条

1 李超;熊璋;朱成军;;基于距离相关图的音频相似性度量方法[J];北京航空航天大学学报;2006年02期

2 苏艳;许南山;;对象持久化技术Hibernate的应用研究[J];福建电脑;2008年03期

3 王红霞;;流媒体中的广告视频检测系统设计[J];光电工程;2010年10期

4 明鸣,朱明;视频流中广告内容的检测[J];计算机工程;2005年06期

5 罗l,

本文编号:2124692


资料下载
论文发表

本文链接:https://www.wllwen.com/wenyilunwen/guanggaoshejilunwen/2124692.html


Copyright(c)文论论文网All Rights Reserved | 网站地图 |

版权申明:资料由用户18178***提供,本站仅收录摘要或目录,作者需要删除请E-mail邮箱bigeng88@qq.com