Loading...

Image Registration by Mapping Defined-Functions Around Regions of Interest

Sarikhani, Hossein | 2013

682 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 44808 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Manzuri, Mohammad Taghi
  7. Abstract:
  8. Image registration is the process of matching two or more images captured at different times, angles or even sensors. Image registration is finding a transformation between two images. Image registration has many applications in different fields of Image Processing and machine vision. In the field of remote sensing, image registration can be used for environmental monitoring, generating graphical maps and gathering data for geogeraphical information systems, also in the field of medical image processing it can be used to detect tumor growth. Generally whenever there is need to extract information from many images; image registration is considered as an important pre-processing step. Different image registration techniques can be divided into two groups: rigid and non-rigid. Non-rigid transfomrmation function can model local deformation and therefore need many parameters. Increase in the number of parameters results in an increase in dimension of transformations search space. Finding a global optimum in a high dimensional space, because of many local optimums, can be very hard. In this thesis an algorithm for non-rigid registration of MRI images of brain has been proposed. This method proposed extracting landmarks based on curvatures in MRI images of the brain. These landmarks and their descriptors are rotation and scale invariant. The proposed registration algorithm is a two step algorithm which in first step extracts landmarks used for initial estimate registration. In this step main structures of brain are matched. In second step images are registered exactly by an area-based technique. Experimental results show the accuracy of proposed method is higher than other well-known methods
  9. Keywords:
  10. Thin Plate Spline ; Image Registration ; Landmark Extraction ; Similarity Measure ; Repeatability Measure ; B-Spline Function ; Magnetic Resonance Imagin (MRI)

 Digital Object List

 Bookmark

No TOC