本文關(guān)鍵詞：電梯監(jiān)控系統(tǒng)的研究與應(yīng)用　出處：《江南大學》2014年碩士論文　論文類型：學位論文

  更多相關(guān)文章： 電梯 智能監(jiān)控 嵌入式系統(tǒng) 異常行為

【摘要】：近年來，隨著高層建筑越來越多，電梯作為最常用的乘坐工具，，與人們的日常生活緊密相連。人們不但要求電梯運行安穩(wěn)、�？烤珳屎统俗孢m，并且對電梯運行的安全性、高效性也提出了更高的要求，于是電梯的安全監(jiān)控成為電梯行業(yè)的重點之一，電梯監(jiān)控系統(tǒng)的研究也成為了一個重要的研究方向。 目前的電梯監(jiān)控系統(tǒng)是由各電梯制造商提供的，功能簡單且開放性較低，這使得電梯難以集中管理和維護。本文以設(shè)計一套功能完善、通用性較好、性能優(yōu)越的電梯監(jiān)控系統(tǒng)為目標，圍繞一些核心問題展開研究，主要研究工作如下： (1)結(jié)合實際情況，對電梯監(jiān)控系統(tǒng)進行需求分析，提出系統(tǒng)總體設(shè)計方案。將電梯監(jiān)控系統(tǒng)劃分為數(shù)據(jù)采集子系統(tǒng)和監(jiān)控中心子系統(tǒng)兩部分，設(shè)計了各個子系統(tǒng)內(nèi)部各個功能模塊的交互協(xié)調(diào)關(guān)系；對數(shù)據(jù)采集子系統(tǒng)進行了軟硬件選型與設(shè)計。 (2)對數(shù)據(jù)采集子系統(tǒng)的設(shè)計進行了詳細研究。硬件采用ARM處理器S3C6410為核心，外擴以太網(wǎng)、圖像采集、加速度傳感器、智能電表以及相關(guān)開關(guān)信號接口，以便采集電梯的相關(guān)數(shù)據(jù)；針對于數(shù)據(jù)采集與傳輸?shù)闹攸c，對視頻采集與傳輸進行了詳細的設(shè)計與分析，包括攝像頭驅(qū)動設(shè)計、視頻圖像數(shù)據(jù)的獲取、圖像編碼壓縮以及網(wǎng)絡(luò)傳輸。 (3)針對電梯轎廂內(nèi)人員安全，為防止斗毆、搶劫等異常行為的發(fā)生，本文提出了一種基于計算機視覺的電梯轎廂內(nèi)異常行為檢測算法，以達到實時安全監(jiān)控的目的。首先，采用混合高斯模型建模提取出運動目標，對運動區(qū)域進行標記；然后統(tǒng)計標記區(qū)域內(nèi)Harris角點的光流信息，將其歸一化成特征矩陣，并建立實時更新的特征矩陣觀測序列；最后利用2維主成分分析(2DPCA)的重構(gòu)原理對觀測序列進行分析，根據(jù)重構(gòu)特征矩陣與原特征矩陣的能量比來判斷是否存在異常行為。通過仿真實驗驗證了算法的有效性。 (4)在設(shè)計的平臺上完成對數(shù)據(jù)采集子系統(tǒng)的開發(fā)，并初步實現(xiàn)整個系統(tǒng)原型，完成了對整個系統(tǒng)的測試工作。 本文設(shè)計的電梯監(jiān)控系統(tǒng)原型對組建區(qū)域性多電梯監(jiān)控、促進電梯監(jiān)控系統(tǒng)標準化有重要的實際意義和應(yīng)用價值。
[Abstract]:In recent years, with the increasing number of high-rise buildings, elevators as the most commonly used riding tool, are closely connected with people's daily life. And the security of elevator operation, high efficiency also put forward higher requirements, so the elevator safety monitoring has become one of the key points of the elevator industry, elevator monitoring system research has become an important research direction. The current elevator monitoring system is provided by the elevator manufacturers, simple function and low openness, which makes it difficult to centralized management and maintenance. The performance of the elevator monitoring system is the goal of the research around some core issues, the main research work is as follows: 1) based on the actual situation, the requirement of elevator monitoring system is analyzed, and the overall design scheme of the system is put forward. The elevator monitoring system is divided into two parts: data acquisition subsystem and monitoring center subsystem. The interaction and coordination relationship of each function module in each subsystem is designed. The software and hardware of the data acquisition subsystem are selected and designed. The hardware uses ARM processor S3C6410 as the core, expanding Ethernet, image acquisition, acceleration sensor. Intelligent electric meter and related switch signal interface, in order to collect elevator related data; Aiming at the emphasis of data acquisition and transmission, the design and analysis of video capture and transmission are carried out in detail, including the design of camera driver, the acquisition of video image data, image coding and compression, and network transmission. In order to prevent the occurrence of abnormal behavior such as brawl and robbery, this paper presents an algorithm for detecting abnormal behavior in elevator car based on computer vision. In order to achieve the purpose of real-time security monitoring. Firstly, the mixed Gao Si model is used to extract the moving target and mark the moving area. Then the optical flow information of the Harris corner in the tagged region is counted and normalized into a feature matrix and a real-time updated observation sequence of the feature matrix is established. Finally, the observation sequence is analyzed by the reconstruction principle of 2-D principal component analysis (PCA). Based on the energy ratio of the reconstructed feature matrix to the original characteristic matrix, the validity of the algorithm is verified by simulation experiments. 4) the data acquisition subsystem is developed on the platform, the prototype of the whole system is realized, and the testing work of the whole system is completed. The prototype of elevator monitoring system designed in this paper has important practical significance and application value in setting up regional multi-elevator monitoring system and promoting the standardization of elevator monitoring system.
【學位授予單位】：江南大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TU857;TP277

【參考文獻】

相關(guān)期刊論文 前9條

1 劉嵐,程莉;嵌入式系統(tǒng)硬件平臺的選型研究[J];電信工程技術(shù)與標準化;2005年02期

2 趙海勇;劉志鏡;張浩;;基于輪廓特征的人體行為識別[J];光電子.激光;2010年10期

3 王憲;慕鑫;宋書林;陳向陽;;視頻序列中基于LBP特征的人體行為識別[J];光電工程;2013年03期

4 王瑋;黃非非;李見為;馮海亮;;采用LBP金字塔的人臉描述與識別[J];計算機輔助設(shè)計與圖形學學報;2009年01期

5 胡芝蘭;江帆;王貴錦;林行剛;嚴洪;;基于運動方向的異常行為檢測[J];自動化學報;2008年11期

6 劉海鵬;廖建新;朱曉民;;PoC中一種負載均衡與時延優(yōu)化的RTP媒體流轉(zhuǎn)發(fā)機制[J];通信學報;2010年08期

7 金亮;張學杰;;3種嵌入式操作系統(tǒng)內(nèi)核的關(guān)鍵技術(shù)分析[J];云南大學學報(自然科學版);2006年S2期

8 白向峰;李艾華;李喜來;李仁兵;;新型背景混合高斯模型[J];中國圖象圖形學報;2011年06期

9 杜鑒豪;許力;;基于區(qū)域光流特征的異常行為檢測[J];浙江大學學報(工學版);2011年07期

相關(guān)博士學位論文 前1條

1 楊霞;高可信嵌入式操作系統(tǒng)體系架構(gòu)研究[D];電子科技大學;2010年

本文編號：1376987