• DocumentCode
    1926569
  • Title

    Comparison of wavelet filters in image coding using hybrid compression technique

  • Author

    Vijayabhaskar, P.V.M. ; Raajan, N.R.

  • Author_Institution
    Sch. of Electr. & Electron. Eng., SASTRA Univ., Thanjavur, India
  • fYear
    2013
  • fDate
    7-9 Jan. 2013
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    In this, we performed compression of an image by using Hybrid (DWT & DCT) technique. Standard wavelet filters of Haar, Daubechies, Symlets, Coiflets, BiorSplines & ReverseBior were used to estimate compression performance. Generally any Compression method is trying to reduce number of bits per pixel for sufficient representation of image. So memory needed for storing necessary information is reduced & communication efficiency is upgraded. In modern days, the method of image decomposition with the help of wavelets has attained an immense agreement of reputation. Totally standard wavelet filters are compared with 3 different gray images in the encoding section & tabulated the MSE Vs PSNR simulation results. These results offered that, Daubechies (db9) wavelet family produced better results with this Hybrid image compression scheme.
  • Keywords
    discrete cosine transforms; discrete wavelet transforms; filtering theory; image coding; image representation; DCT; DWT; MSE; PSNR; communication efficiency; compression performance; gray images; hybrid compression technique; image coding; image decomposition; image representation; standard wavelet filters; Biomedical imaging; Discrete cosine transforms; Encoding; Image coding; PSNR; Prototypes; Arithmetic coding; Discrete Cosine Transform; Discrete Wavelet Transform;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Emerging Trends in VLSI, Embedded System, Nano Electronics and Telecommunication System (ICEVENT), 2013 International Conference on
  • Conference_Location
    Tiruvannamalai
  • Print_ISBN
    978-1-4673-5300-7
  • Type

    conf

  • DOI
    10.1109/ICEVENT.2013.6496570
  • Filename
    6496570