skip to main content
資源種類 顯示結果: 顯示結果: 查詢種類 索引

植基於多目標基因演算法之影像浮水印系統

莊鳳屏; Chuang, Feng-ping 李建樹;黃意雯; Jiann-Shu Lee;I-Wen Huang; 數位學習科技學系碩士班 2009

線上取得

  • 題名:
    植基於多目標基因演算法之影像浮水印系統
  • 著者: 莊鳳屏; Chuang, Feng-ping
  • 李建樹; 黃意雯; Jiann-Shu Lee; I-Wen Huang; 數位學習科技學系碩士班
  • 主題: 浮水印; 基因演算法; 多目標基因演算法; Watermark; Genetic Algorithm; Multi-objective Genetic Algorithm
  • 描述: 本論文提出一植基於多目標基因演算法之影像浮水印系統,此系統以多目標最佳化策略,設計出兼具良好影像品質與高強健性的浮水印嵌入法。此浮水印系統相較於傳統最佳化式的浮水印演算法,有三個重要的優點。 首先,使用多目標基因演算法能產生一組最佳解,而非單一個最佳解,因此,使用者可從解集合中挑選想要的最適解,以類似客製化的方式滿足不同使用者的需求。其次,運用多目標基因演算法所產生的解集合,其穩定性比起傳統最佳化式演算法高出很多。再者,以多目標基因演算法所求得的解,不論就影像品質抑或是印強健性,均能優於傳統最佳化式浮水印演算法所得的結果。而一系列的實驗結果證實,本系統相較於傳統最佳化式浮水印演算法,具有上述之三個重要優點。
    In this thesis, we propose a novel image watermarking system based on multi-objective genetic algorithm (MOGA), that can achieve both good image quality and high robustness. Compared with traditional optimization-based algorithms, this system possesses three major superiorities. First, the proposed method can generate a set of optimal solutions rather than a single one. Hence, the user can pick out one optimal solution best fit for his application. That is, our method can provide the user multiple optimal solutions to achieve approximate customization for image watermarking. Next, the solutions generated by MOGA are more stable than those by traditional optimization-based algorithms. Third, the solutions derived by MOGA can superior to those by traditional optimization-based algorithms for both image quality and robustness. The experimental results also reveal that the proposed image watermarking system possesses the above mentioned three superiorities.
    碩士
  • 建立日期: 2009
  • 格式: 121 bytes
    text/html
  • 語言: 中文
  • 識別號: http://nutnr.lib.nutn.edu.tw/handle/987654321/2915
  • 資源來源: NUTN IR

正在檢索遠程資料庫,請稍等

  • 查詢:
  • scope:("NUTN"),scope:(NUTN_ALEPH),scope:(NUTN_IR),scope:(NUTN_SFX),primo_central_multiple_fe
  • 顯示現有記錄