English  |  正體中文  |  简体中文  |  Items with full text/Total items : 49279/83828 (59%)
Visitors : 7146468      Online Users : 80
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/76909


    Title: 應用於高安全性保全智慧型多攝影機監視系統之研究與實現
    Other Titles: Research and Implemertation of an Intelligent Multi-Camera System for High Security Surveillance
    Authors: 江正雄
    Contributors: 淡江大學電機工程學系
    Keywords: 智慧型監控系統;分散式視訊編碼技術;多攝影機協同設計;移動物件偵測及追蹤模組;快速影像視訊檢索與分類模組;身份辨識模組;步態辨識;遠距離人臉辨識;Intelligent surveillance system;multi camera co-operation;video abstraction and classification;human being identification;human gait recognition
    Date: 2011-08
    Issue Date: 2012-05-22 21:51:54 (UTC+8)
    Abstract: 子計畫四名稱- 應用於高安全性保全智慧型多攝影機監視系統之研究與實現 為了保障生命安全與財產穩固,保全(Surveillance)自古以來一直就倍受重視,其技術也一直在更新一直在進步。自從攝影機的發明,尤其是數位式及高解析度監視攝影機的發明與普及後,保全監控的發展更進入了一個新的紀元,人們可以藉著架設監視攝影機於關鍵位置,對該處仔細的監控。然而,大量的監視攝影資訊管理並不容易,需要花費大量的人力來監視畫面,容易造成人為疏失,而導致安全維護效率的降低。因此,為了解決人為上的疏忽,必須發展一套智慧型攝影監控保全系統,來取代傳統的人力監控,以提升監視效率。有鑑於此,我們乃提出一個兩年期的智慧型多攝影機監控保全系統研究計畫,希望應用於高安全性的場所如博物館等。此研究計畫為架構於「以分散式視訊編碼技術為主之非可破壞無線視訊安全監控系統之研究」總計畫下的一個子計畫,本智慧型多攝影機監控保全系統研究計畫包含三個子題模組,分別為:多攝影機協同與移動物件偵測及追蹤模組、快速影像視訊檢索與分類模組、與身份辨識模組。多攝影機協同與移動物件偵測及追蹤主要研究如何利用影像,並結合多攝影機協同來作即時的物件偵測與物件追蹤,並記錄物件的行走路線與軌跡;快速影像視訊檢索與分類的部分則應用Video Synopsis的技術,來作視訊檢索以及更進一步作物件分類;而身份辨識主要研究遠距離人臉辨識及步態辨識。我們希望在兩年內,把這幾個模組加以研究並整合,最終成為一個可行的智慧型監控保全系統。這三個模組的各年度研究內容如下所述: 1. 多攝影機協同與移動物件偵測及追蹤模組 第一年:室內的物件辨識與追蹤,並利用降低影像維度以達到多攝影機即時處理之目的。 第二年:多攝影機協同應用於行走路線的追蹤。 2. 快速影像視訊檢索與分類模組 第一年:Video Synopsis的影像視訊檢索技術研究。 第二年:Video Synopsis的資料分類技術研究及軟硬體共同設計。 3. 身份辨識模組 第一年:步態辨識演算法研究。 第二年:遠距離人臉辨識演算法研究。
    Research and Implementation of an Intelligent Multi-Camera System for High Security Surveillance For securing the life and property safety, people pay a lot of attention on surveillance even from the ancient time. Because of the safety matter, the technique for surveillance kept improving all the time. After the invention of monitoring camera, especially the invention of the digital and high resolution monitoring camera, the development of surveillance steps forward a very big pace. People usually use monitoring cameras to monitor whatever they like. However, the management of the monitoring information becomes more difficult and sometimes people are overwhelmed by the large amount of the monitoring information. For better surveillance quality, the security department even has to hire more people for handling the monitoring information. The manual operation usually makes carelessness and it may cause security problems. Therefore, it needs to develop an intelligent monitoring surveillance system to replace the traditional manual operation system to improve the monitoring efficiency. For a good monitoring consideration in the surveillance system, we are proposing a two-year planning research project for intelligent multi camera monitoring surveillance system. We expect this intelligent surveillance system can be applied to a highly secure environment such as a public museum. This research planning project is a sub-project of the main project: “Indestructible Video Surveillance System with New Low-Cost Distributed Video Coding (DVC) Paradigm over Wireless Network”. This intelligent multi camera monitoring surveillance project consists of three parts (modules) and they are: (1) multi camera co-operation and objects recognition and tracking module, (2) high speed video abstraction and classification module, and (3) human being identification module. The multi camera co-operation and objects recognition and tracking module tries to use the multi camera scheme to finish the task of object recognition and tracking. Further it tries to record the moving route and trajectory of the object. The high speed video abstraction and classification module uses the Video Synopsis technique to finish the task of video abstraction and further to accomplish the video classification. The human being identification module tries to identify the human face from long distance and human gait. We expect to integrate these three modules to an intelligent monitoring surveillance system in two years. This planning project is a two-year research project, and the contents of the modules that we expect to accomplish are described year by year as follows:
    Appears in Collections:[電機工程學系暨研究所] 研究報告

    Files in This Item:

    There are no files associated with this item.

    All items in 機構典藏 are protected by copyright, with all rights reserved.


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback