• DocumentCode
    446764
  • Title

    Using AM-FM image modeling technique in mammograms

  • Author

    Elshinawy, M.Y. ; Chouikha, M.F.

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Howard Univ., Washington, DC, USA
  • Volume
    2
  • fYear
    2003
  • fDate
    27-30 Dec. 2003
  • Firstpage
    660
  • Abstract
    This paper represents preliminary results for a full study that tries to answer the following question, how well AM-FM modeling technique will behave when applied to mammograms. Our results thus far show that the AM-FM modeling technique yields good results when the images used are mammogram The AM-FM modeling technique consists of three main steps. The first step is applied in this paper, which is, passing the image under study through a bank of Gabor filters. These filters are band pass filters. We used 40 filters arranged as eight rays with 5 filters per ray. Each with has its angle with different center frequencies. The conducted experiments show that by adding up the filtered image from all the filters in the bank, the original image can be observed. This means that Gabor filter banks can work for mammograms and the future work will concentrate on applying the suitable AM-FM modeling technique to detect the region of interest on mammograms.
  • Keywords
    Gabor filters; band-pass filters; mammography; medical image processing; AM-FM image modeling; Gabor filters; band pass filters; filter bank; mammograms; Band pass filters; Breast biopsy; Breast cancer; Cancer detection; Filter bank; Frequency conversion; Gabor filters; Image analysis; Image texture analysis; Mammography;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Circuits and Systems, 2003 IEEE 46th Midwest Symposium on
  • ISSN
    1548-3746
  • Print_ISBN
    0-7803-8294-3
  • Type

    conf

  • DOI
    10.1109/MWSCAS.2003.1562373
  • Filename
    1562373