國立臺灣大學電機工程學系暨研究所貝蘇章2006-07-252018-07-062006-07-252018-07-062004-07-31http://ntur.lib.ntu.edu.tw//handle/246246/7988在影像處理上，一般而言，不同的影像在頻域當 中會有不同的振幅及不同的相角。然而，在這裡，我 們發現了，只要將原影像加上了適當的虛數部分，就 可以將這些影像的振幅頻譜變得相同而只有相角頻譜 不同。或者，反過來，將它們的相角頻譜變相同而只 有振幅頻譜不同。我們稱前者為相位金匙演算法，稱 前者為振幅金匙演算法。這些演算法可適用於分數傅 氏轉換，線性完整轉換，及分數正絃餘絃轉換等等。 它們對於影像或聲波處理，提供了一種有效的加密保 密及編碼的方式。Different images always have different amplitude spectrums. However, in this paper, we show that, with the appending of proper imaginary parts, we can make a series of natural images have the same amplitude spectrum but different phase key in the frequency domain. Since the only difference among the spectra is phase, we can use the phase spectrum to represent each of the images. Different images correspond to different phase keys. It is useful for image encryption and data compression. Our algorithm can be applied for not only the FT but also the fractional Fourier transform (FRFT) and many other operators. Since many parts of the algorithm can be treated as keys, we can effectively encrypt an image. Moreover, we can also reverse the direction of the algorithm, i.e., make a series of images have the same phase spectrum and use the amplitude spectra as the keys to distinguish different images.application/pdf304286 bytesapplication/pdfzh-TW國立臺灣大學電機工程學系暨研究所相位金匙振幅金匙離散分數傅氏轉換離散線性完整轉換phase key and amplitude key algorithmsdiscrete fractional Fourier transformdiscrete linear canonical transformreporthttp://ntur.lib.ntu.edu.tw/bitstream/246246/7988/1/922213E002089.pdf