移动终端中主动预警系统的设计与实现

发布时间：2018-04-01 01:32

本文选题：Android平台　切入点：智能监控　出处：《江苏科技大学》2017年硕士论文

【摘要】：随着4G网络的普及,智能手机已经是日常生活中不可缺少的一部分,同时安防监控也是日常生活中值得关注的问题。为了改善传统视频监控系统主动预警功能不强、避免在事后取证时观看大量的无效静止场景的视频录像,实现监控时达到主动预警的目的,本文设计了一个基于移动终端的主动预警系统。本文根据实际监控场景的多样化,对运动目标检测采用背景建模的方法,同时增加场景中人脸检测。针对实际问题,研究了有关文献,介绍了与智能监控相关的方法和技术。本文研究了采用背景差分法、光流法、帧差法对运动目标进行检测的方法和技术;研究了PCA算法、Fisher脸法、Adaboost算法和Haar级联中的boosting算法等人脸识别算法。在视屏传输算法方面,主要研究了H.264的RTP封包算法以及其具体过程。为了在移动终端的实现主动预警,本文在Android平台下,按照软件工程的设计的规范,对移动终端中主动预警系统进行了需求分析、系统设计和实现。本项目实现的主动预警系统包括服务器端服务程序和移动终端的应用程序。服务器端实现了有场景变化时就录制视频并保存,在主动预警方面,提出分级预警策略,使得系统减少误报,保障了系统的可用性,使得用户在移动端接收到报警信息后,根据报警级别采取相应措施,及时地处理。在研究的过程中,还针视频监控传输视频数据过程中数据量大的问题,对H.264算法进行了研究和改进,使得本项目研究的主动预警系统实时性得到了提高。另外,为了对视屏监控中的入侵对象进行区分,本文对人脸识别算法行了研究,通过人脸识别来辅助判断的入侵对象,提高主动预警准确率,减少误报的概率。本课题通过对运动目标检测、人脸检测及视频传输技术的研究,研发实现了一个基于移动终端的主动预警系统。
[Abstract]:With the popularity of 4G network, smart phone is an indispensable part of daily life, and security monitoring is also a problem worth paying attention to in daily life.In order to improve the traditional video surveillance system active warning function is not strong, avoid watching a large number of video videos of invalid static scene after evidence collection, and achieve the purpose of active early warning when monitoring.An active early warning system based on mobile terminal is designed in this paper.According to the diversity of the actual monitoring scene, the background modeling method is used to detect moving targets, and the face detection in the scene is added at the same time.In view of the practical problems, the related literature is studied, and the methods and techniques related to intelligent monitoring are introduced.In this paper, the background difference method, optical flow method and frame difference method are used to detect moving targets, and face recognition algorithms such as PCA algorithm and boosting algorithm in Haar cascade are studied.In video transmission algorithm, H.264 RTP packet algorithm and its specific process are studied.In order to realize active early warning in mobile terminal, according to the design standard of software engineering, the requirement analysis, system design and implementation of active early warning system in mobile terminal are carried out in this paper based on Android platform.This project realizes the active early warning system including server-side service program and mobile terminal application program.The server can record the video and save the video when the scene changes. In the aspect of active early warning, a hierarchical warning strategy is put forward, which can reduce the false positives, ensure the usability of the system, and make the user receive the alarm information in the mobile terminal.Take appropriate measures according to the alarm level and deal with them in time.In the process of research, the problem of large amount of data in video surveillance and transmission process is discussed. The H.264 algorithm is studied and improved, which improves the real-time performance of the active early warning system studied in this project.In addition, in order to distinguish the intrusion objects in video surveillance, this paper studies the face recognition algorithm, through face recognition to assist the judgment of the intrusion object, improve the accuracy of active early warning, reduce the probability of false positives.Based on the research of moving target detection, face detection and video transmission, an active early warning system based on mobile terminal is developed.
【学位授予单位】：江苏科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41;TP277

【参考文献】