Web-Books
in the Austria-Forum
Austria-Forum
Web-Books
International
Proceedings - OAGM & ARW Joint Workshop 2016 on "Computer Vision and Robotics“
Page - 17 -
  • User
  • Version
    • full version
    • text only version
  • Language
    • Deutsch - German
    • English

Page - 17 - in Proceedings - OAGM & ARW Joint Workshop 2016 on "Computer Vision and Robotics“

Image of the Page - 17 -

Image of the Page - 17 - in Proceedings - OAGM & ARW Joint Workshop 2016 on

Text of the Page - 17 -

Directional  Wavelet  based  Features  for  Colonic  Polyp Classification Georg  Wimmer1,  Michael  Häfner3,  Shigeto  Joshida4,  Toru  Tamaki5,  Shinji  Tanaka4, Jens  Tischendorf2  and  Andreas  Uhl1 1  University  of  Salzburg,  Austria;  2  RWTH  Aachen  University  Hospital;  3  St.  Elisabeth  Hospital;  4  Hiroshima  University  Hospital;  5  Hiroshima  University Abstract In  this  work,  various  wavelet  based  methods  like  the  discrete  wavelet  transform,  the  dual­ tree complex  wavelet  transform,  the  Gabor  wavelet  transform,  curvelets,  contourlets  and  shearlets  are applied  for  the  automated  classification  of  colonic  polyps.  The  methods  are  tested  on  8  HD­ endoscopic  image  databases,  where  each  database  is  acquired  using  different  imaging  modalities (Pentax's   i­ Scan  technology  combined  with  or  without  staining  the  mucosa),   2  NBI  high­ magnification  databases  and  one  database  with  chromoscopy  high­ magnification  images.  To evaluate  the  suitability  of  the  wavelet  based  methods  with  respect  to  the  classification  of  colonic polyps,  the  classification  performances  of  3  wavelet  transforms  and  the  more  recent  curvelets, contourlets  and  shearlets  are  compared  using  a  common  framework.  Wavelet  transforms  were already  often  and  successfully  applied  to  the  classification  of  colonic  polyps,  whereas  curvelets, contourlets  and  shearlets  have  not  been  used  for  this  purpose  so  far.  We  apply  different  feature extraction  techniques  to  extract  the  information  of  the  subbands  of  the  wavelet  based  methods. Most  of  the  in  total  20  approaches  were  already  published  in  different  texture  classification contexts.  Thus,  the  aim  is  also  to  assess  and  compare  their  classification  performance  using  a common  framework.  Three  of  the  20  approaches  are  original.  These  three  approaches  extract Weibull  features  from  the  subbands  of  curvelets,  contourlets  and  shearlets.  Additionally,  5  state­ of­ the­ art  non  wavelet  based  methods  are  applied  to  our  databases  so  that  we  can  compare  their results  with  those  of  the  wavelet  based  methods.  It  turned  out  that  extracting  Weibull  distribution parameters  from  the  subband  coefficients  generally  leads  to  high  classification  results,  especially for  the  dual­ tree  complex  wavelet  transform,  the  Gabor  wavelet  transform  and  the  Shearlet transform.  These  three  wavelet  based  transforms  in  combination  with  Weibull  features  even outperform  the  state­ of­ the­ art  methods  on  most  of  the  databases.  We  will  also  show  that  the Weibull  distribution  is  better  suited  to  model  the  subband  coefficient  distribution  than  other commonly  used  probability  distributions  like  the  Gaussian  distribution  and  the  generalized Gaussian  distribution. 17
back to the  book Proceedings - OAGM & ARW Joint Workshop 2016 on "Computer Vision and Robotics“"
Proceedings OAGM & ARW Joint Workshop 2016 on "Computer Vision and Robotics“
Title
Proceedings
Subtitle
OAGM & ARW Joint Workshop 2016 on "Computer Vision and Robotics“
Authors
Peter M. Roth
Kurt Niel
Publisher
Verlag der Technischen Universität Graz
Location
Wels
Date
2017
Language
English
License
CC BY 4.0
ISBN
978-3-85125-527-0
Size
21.0 x 29.7 cm
Pages
248
Keywords
Tagungsband
Categories
International
Tagungsbände

Table of contents

  1. Learning / Recognition 24
  2. Signal & Image Processing / Filters 43
  3. Geometry / Sensor Fusion 45
  4. Tracking / Detection 85
  5. Vision for Robotics I 95
  6. Vision for Robotics II 127
  7. Poster OAGM & ARW 167
  8. Task Planning 191
  9. Robotic Arm 207
Web-Books
Library
Privacy
Imprint
Austria-Forum
Austria-Forum
Web-Books
Proceedings