Image Compression using Hartley transform (bibtex)
by Ahmed Nabil Belbachir, Ted Chilton, Marc Nunkesser, Sahbi Sidhom, Georges Szajnowski
Abstract:
The report presents a novel approach for image compression using the Hartley transform (HT). The Hartley transform has the advantage of solving the problem of phase wrapping from which the Fourier transform su.ers. The magnitude and phase compression using this transformation (HT) have proved better performance than those of the Fourier Transform. Magnitude and phase were processed separately. The quantization of frequency samples in less bits has increased the compression ratio.Furthermore, the distributions used to generate the noise signi.cantly in.uence the result. The lossy compression technique used seems not to degrade the image quality. A nonlinear filter for smoothing the resulting image would be suitable for image enhancement. In general, the overall compression ratio is acceptable it compresses to about $15-30$ percent the size of the original image. A lossless compression technique could be performed additionally to increase the compression factor.
Reference:
Image Compression using Hartley transform (Ahmed Nabil Belbachir, Ted Chilton, Marc Nunkesser, Sahbi Sidhom, Georges Szajnowski), Technical report, PRIP, TU Wien, 2003.
Bibtex Entry:
@TechReport{TR087,
  author =	 "Ahmed Nabil Belbachir and Ted Chilton and Marc
                  Nunkesser and Sahbi Sidhom and Georges Szajnowski",
  title =	 "Image {C}ompression using {H}artley transform",
  institution =	 "PRIP, TU Wien",
  number =	 "PRIP-TR-087",
  year =	 "2003",
  url =		 "ftp://ftp.prip.tuwien.ac.at/pub/publications/trs/tr87.pdf",
  abstract =	 "The report presents a novel approach for image
                  compression using the Hartley transform (HT). The
                  Hartley transform has the advantage of solving the
                  problem of phase wrapping from which the Fourier
                  transform su.ers. The magnitude and phase
                  compression using this transformation (HT) have
                  proved better performance than those of the Fourier
                  Transform. Magnitude and phase were processed
                  separately. The quantization of frequency samples in
                  less bits has increased the compression
                  ratio.Furthermore, the distributions used to
                  generate the noise signi.cantly in.uence the
                  result. The lossy compression technique used seems
                  not to degrade the image quality. A nonlinear filter
                  for smoothing the resulting image would be suitable
                  for image enhancement. In general, the overall
                  compression ratio is acceptable it compresses to
                  about $15-30$ percent the size of the original
                  image. A lossless compression technique could be
                  performed additionally to increase the compression
                  factor.",
}
Powered by bibtexbrowser