English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 49378/84106 (59%)
造訪人次 : 7377776      線上人數 : 71
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: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/35895

    題名: Log-Gabor小波為基礎之人臉合成系統
    其他題名: Facial image aging synthesis based on log-gabor wavelet
    作者: 李祐丞;Lee, Yu-cheng
    貢獻者: 淡江大學電機工程學系碩士班
    謝景棠;Hsieh, Ching-tang
    關鍵詞: 人臉偵測;Log-Gabor 小波;AdaBoost演算法;年齡合成;Face detection;Log-Gabor wavelet;Adaboost algorithm;Aging synthesis
    日期: 2009
    上傳時間: 2010-01-11 07:17:20 (UTC+8)
    摘要: 近年來,已出現了許多臉部影像合成之應用,例如:視覺娛樂、電影動畫製作、遊戲、化妝品測試、找尋失踪人口…等。而在這當中,為了找尋失踪人口進而預測出不同年齡之人臉合成,便為我們所期望努力的方向。不同年齡人臉的合成操作成為了現今熱門的話題之一。
    At present, there are many application of facial image synthesis by computer graphics such as visual entertainment, movie production, game, cosmetic examinations, and searches for missing persons, etc. For these reason, manipulation of different age on human face become a common problem and the most popular one nowadays.

    In this paper, we propose an approach for manipulating different age texture on human face base on face detection, and Log-Gabor wavelet. First of all, Adaboost algorithm is used to get the main facial features and to normalize them into the same size and same features position of each picture. Second, we find an appropriate target age image which is similar to subject’s image, and then we employ the properties of multi-resolution and multi-channel to extract the decomposition maps of age texture with different age on human face by using Log-Gabor wavelet. Finally, we can effectively manipulate the different age of human face through controlling the amount of decomposition map of target age images to cover on the subject’s faces image.

    Experimental results show that the aging synthesis of facial images can be generated well by using our proposed approach.
    顯示於類別:[電機工程學系暨研究所] 學位論文


    檔案 大小格式瀏覽次數



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