Loading...

Automatic Analysis and Tracking of Motile Cells in Video Microscopy

Shayegh, Zahra | 2011

1166 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 41571 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Vosughi Vahdat, Bijan; Rabiei, Hamid Reza; Salman Yazdi, Reza
  7. Abstract:
  8. Analysis of semen and quality assessments of sperm cells is of great importance in scrutiny of male fertility. Several methods have been introduced for analyzing and identification of the sperm motility and morphology in a semen sample. Identifying and tracking of rapid and variant movements of multiple sperms, in a short duration of time, is somewhat difficult and complex for human, even for an expert. Then applying semi-automated or automated (un-supervised) methods, based on image analysis and computing, spread fast and computer aided semen analyzer systems, became wildly used in clinical and research laboratories.
    In this paper we propose an efficient multiple tracking methods to track sperms in video microscopy images and handle occlusion along the sequence. Algorithm is based on the detection of target in each frame, and then the tracker searches for every point’s correspondence in sequential frames. We provide quantitative evaluations of the proposed method against existing common particle method and commercial sperm analyzer software and the provided ground truth. Experimental results demonstrate that our approach truly corresponds the objects and is robust against occlusion and misdetection.
  9. Keywords:
  10. Image Processing ; Object Tracking ; Particle Filter ; Mean Shift ; Kernel Learning ; Sperm Track ; Infertility ; Semen Analysis ; Computer Aided Semen Analyzer (CASA) ; Points Correspondence ; Silhouette

 Digital Object List

  • محتواي پايان نامه
  •   view

 Bookmark

No TOC