English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 2928/5721 (51%)
造訪人次 : 375847      線上人數 : 561
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU 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://ir.lib.ncut.edu.tw/handle/987654321/6542


    題名: Transferring color to Grayscale Images Using Vector Quantization of Luminance Mapping Technique
    作者: 陳文淵
    貢獻者: 電子工程(學)系
    日期: 2009-02
    上傳時間: 2017-11-01 09:34:48 (UTC+8)
    摘要: We develop an effective algorithm for colorizing a grayscale image. In our approach, a reference color image, an RGB to Lαβ color transform L=luminance, α β=chrominance), and a block-based vector quantization of luminance mapping (VQLM) technique are used to automatically colorize the grayscale image. The VQLM technique compares the grayscale image with the luminance of the color reference image to obtain the information of the αβ planes of the grayscale image. After the chrominance is padded, the inverse color transform, L αβ to RGB, colorizes the grayscale scene is colorized. Meanwhile, we create a mean of VQLM (MVQLM) method to improve the quality of the colorized grayscale image. Experimental results show the MVQLM method is better than the VQLM method. Also, we investigate colorizing the grayscale image working in the Lαβ and YIQ spaces. The simulation results also reveal that working in the Lαβ space is slightly better than working in the YIQ space. Compared to other colorizing schemes, our proposed method has two advantages: 1. the codebook and MVQLM techniques colorize the grayscale images for any size image that can be evenly divided by 2 automatically; and 2. the MVQLM method obtains a smoother colorizing effect and improved quality compared to the VQLM method.
    關聯: Optical Engineering
    顯示於類別:[電子工程系(所)] 【電子工程系所】期刊論文

    文件中的檔案:

    沒有與此文件相關的檔案.



    在NCUTIR中所有的資料項目都受到原著作權保護.


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