English  |  正體中文  |  简体中文  |  Items with full text/Total items : 49942/85109 (59%)
Visitors : 7785091      Online Users : 45
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/987654321/74573


    Title: 在雲端計算中使用支援向量機來做性別分類
    Other Titles: Using support vector machines for gender classification on cloud computing
    Authors: 周俊昌;Chou, Chun-Chang
    Contributors: 淡江大學資訊工程學系碩士在職專班
    洪文斌;Horng, Wen-Bing
    Keywords: 支援向量機;區域二元模式;性別辨識;雲端計算;Cloud Computing;Gender classification;LBP;SVM
    Date: 2011
    Issue Date: 2011-12-28 18:55:34 (UTC+8)
    Abstract: 在近幾年電腦視覺影像研究中,人臉的辨識上已有相當大的進步,延伸而來的應用也相當多,如人臉的性別、表情、年齡辨識等。在這些影像的辨識研究領域中,系統的辨識率以及系統的效能,一直是相關領域的研究重點。
    在本論文中,輸入影像可經由個人電腦選取,或經由手機端擷取數位相機影像。經由用戶端裝置偵測人臉的位置,並使用區域二元模式演算法作為人臉正面影像的特徵值,以及降低此特徵值維度以減低系統的運算負擔,之後傳送至伺服器端以辨識性別。在性別辨識的問題中,我們利用支援向量機的統計學理論來做為人臉的性別辨識分類器。
    支援向量機演算法對於在計算能力或記憶體容量比較低裝置是相當大的負擔如個人掌上型電腦或手機等。為了解決這類的問題,我們開發了主從式架構的程式,在伺服器端負責訓練性別分類器,及對用戶端所傳送之影像特徵值辨識性別並把辨識的結果回報給用戶端的裝置,以避免用戶者端的裝置運算負荷過大。在本實驗中,我們對FERET人臉資料庫做交叉驗證,可得到93.21%辨識率。
    Face detection has been considerable progress in computer vision recently, and extends from the applications are quite a few, such as gender recognition, expression, age identification. In the field of computer vision, recognition rate and system performance has been attention in the related field.
    In this paper, client device is responsible for detecting the location of face then reduce dimension of the face feature using LBP (local binary patterns) operator, and send to server side. The classification problem is performed by support vector machines.
    The support vector machine algorithm requests amount of computing and memory, which is a considerable burden on handheld computers or mobile phones. To solve the problem, we developed client-server programming. In order to avoid that client device over loading, the server side is responsible for training of gender classifier and predicts results which received the face feature form client device then sends to the client device. In our experiments, 5-fold cross-validation are performed on FERET face database and obtained the accuracy of 93.21%.
    Appears in Collections:[資訊工程學系暨研究所] 學位論文

    Files in This Item:

    File SizeFormat
    index.html0KbHTML125View/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