淡江大學機構典藏:Item 987654321/52539
English  |  正體中文  |  简体中文  |  Items with full text/Total items : 62830/95882 (66%)
Visitors : 4127002      Online Users : 408
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
    Home ‧ Login ‧ Upload ‧ Help ‧ About ‧ Administer Goto mobile version


    Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/52539


    Title: 即時人臉偵測之軟硬體共同設計
    Other Titles: Hardware/software codesign of real-time human face detection
    Authors: 余家潤;Yu, Chia-jun
    Contributors: 淡江大學電機工程學系碩士班
    翁慶昌;Wong, Ching-chang
    Keywords: 人臉偵測;軟硬體共同設計;SoPC;影像處理電路;Face Detecting;Hardware/Software Codesign;SOPC
    Date: 2010
    Issue Date: 2010-09-23 17:54:15 (UTC+8)
    Abstract: 本論文提出一個在多主從系統架構下以軟硬體共同設計的方法來實現即時的人臉偵測。本論文使用一個可以軟硬體共同設計的SOPC (System On a Programmable Chip)開發平台來進行即時影像處理電路之設計，其設計方式是以多主從系統架構為基礎，使用軟體及硬體共同設計的方式實現人臉偵測主要的五個步驟：(1) 影像二值化(Image Binary)，(2) 影像增強處理(Image Enhancement)，(3) 邊緣化(Edge Detection)，(4) 物件分割(Object Segmentation)，及(5) 人臉特徵比對(Feature Comparison)。在軟硬體共同設計的劃分上，由於影像二值化、影像增強處理及邊緣化等前三個步驟屬於影像前處理的部分，並且系統資源消耗較高，所以將在FPGA晶片內以硬體描述語言(HDL)來建置電路的硬體方式實現，並搭配多個主控端來提升影像擷取、顯示及儲存影像處理結果的速度。而相較系統資源消耗較低的物件分割及人臉特徵比對這兩個步驟，則會在Nios II處理器內以撰寫C語言的軟體方式實現。本論文以Verilog所實現的硬體電路來取代一般由C語言軟體實現的影像處理，使得整體系統在SOPC軟硬體共同設計下具有更佳的即時影像處理的效能。在實際驗證上，本論文是以友晶科技所發展的多媒體開發板DE2-70為實驗平台，從實驗結果可知所提之架構與方法確實可以用較少的影像處理時間，達到即時的人臉偵測。
    In this thesis, a Hardware/Software (HW/SW) codesigned method based on a multiple master-slave system architecture is proposed to implement a real-time human face detection. In the design and implementation of real-time human face detection, a software and hardware codesign method which integrates C language and Hardware Description Language (HDL) based on a SOPC (System on a Programmable Chip) technique is applied to design five image processing modules: (1) Image Binary, (2) Image Enhancement, (3) Edge Detection, (4) Object Segmentation, and (5) Feature Comparison. In order to get real-time image processing, these modules which cost more computing time are implemented by using hardware accelerating circuits. Then, in the multiple master-slave system architecture, multiple master units are design to accelerate the process speed of image capture and display. Some experiment results illustrate that the image processing time is reduced effectively by the proposed method at the DE2-70 development board so that a real-time human face detection can be implemented.
    Appears in Collections:[Graduate Institute & Department of Electrical Engineering] Thesis

    Files in This Item:

    File SizeFormat
    index.html0KbHTML323View/Open

    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