淡江大學機構典藏:Item 987654321/35840
English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 62830/95882 (66%)
造訪人次 : 4150018      線上人數 : 631
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋
    請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/35840


    題名: 應用水平邊緣特徵於人臉偵測系統
    其他題名: Application to face detection using horizental edge features
    作者: 陳中興;Chen, Chung-hsing
    貢獻者: 淡江大學電機工程學系碩士在職專班
    賴友仁;Lai, Eugene
    關鍵詞: 小波轉換;邊緣檢測;條列式規則區分法;特徵區分法;the wavelet transform;edge detection;knowledge-based method;feature-based method
    日期: 2006
    上傳時間: 2010-01-11 07:13:24 (UTC+8)
    摘要: 人臉偵測對於「臉孔辨識」、「駕駛睡眠偵測」與「智慧型人機介面」等應用是非常重要的,最近幾年來更是受到大家的廣泛重視。這些應用的首要工作在於找出人臉的所在區域,因此如何正確而可靠的找出人臉位置是相當重要的。在本論文提出了一個新的 大小的邊緣遮罩,同時利用可調式閥值的調整,對輸入的影像進行特徵邊緣檢測,可以突顯出欲偵測之人臉眼部、鼻孔及嘴部的水平方向性邊緣特徵。首先將影像利用小波轉換消除雜訊及降低解析度,由本文提出的新的邊緣檢測方法,建立臉部的水平方向性邊緣特徵,再利用多層次特徵模板進行眼部及嘴部的特徵比對,進而將正確的人臉位置標示出來。
    Human face detection is very important in its wide range of applications, such as human face recognition, drowsiness detection, human-computer interface, etc. In most such applications, the existence of human faces and their correct locations must be found. In other words, a reliable and correct method for detecting and locating human face region is considerably significant. In this paper, a new 1x7 mask for edge detection combining the adjustment of adapted threshold values is proposed. In this way, we can increase the edge orientation features of eyes and mouths. First, the wavelet transform is used to reduce the noise and the dimension. Second, a new 1x7 mask is employed to build the horizental edge features. Third, a multilayer face template is used to match the horizental edge features of faces. Finally, a reliable and correct method is developed to locate human face region.
    顯示於類別:[電機工程學系暨研究所] 學位論文

    文件中的檔案:

    檔案 大小格式瀏覽次數
    0KbUnknown307檢視/開啟

    在機構典藏中所有的資料項目都受到原著作權保護.

    TAIR相關文章

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