You are here:
Publication details
Fast Anisotropic Filtering and Performance Evaluation Tool for Optical Flow in Biomedical Image Analysis
Authors | |
---|---|
Year of publication | 2011 |
MU Faculty or unit | |
Citation | |
Description | The thesis is focused on the analysis of time-lapse images acquired using a fluorescence light microscope. In particular, for the purpose of automated evaluation of motion of stained cell structures, e.g., proteins or cell nuclei, perceived over a time period, we aim towards an object tracking based on an optical flow field. An optical flow method estimates a flow field in which a vector is assigned to every pixel in an image. The vector represents the difference in position of the same pixel content between two images. To track the given position it is then enough to simply follow flow vectors provided good flow estimates are available. The thesis reviews the process from acquiring image data to methods for computing optical flow. The description starts with the limits of the imaging technology and characterization of the obtained image data. The survey part reviews and discusses methods that allow for conducting object tracking. |
Related projects: |
|