The development of a quantitative phase microscope (QPM) has allowed the ability to acquire real-time phase movies of biological processes. The image processing of the data is critical to the system's ability to measure relative changes. The phase data must be consistent throughout a measurement and background fluctuations must be minimized. The research presented in this work discusses methods to effectively process sequences of phase data such that it can be used to quantify changes within real-time studies of living cells. This work begins by exploring two-dimensional phase unwrapping to determine the most effective ways to estimate the measured phase surface. Conventional methods of comparing unwrapping performance will be used. In addition, a novel method will be introduced that can characterize accuracy using continuity of derivatives. It will be shown that the most accurate phase estimates are made using modulation data with quality-guided phase unwrapping. After two-dimensionally unwrapping all frames of data within a measurement, there are background fluctuations due to residual surface shape as well as mean phase value fluctuations. Traditionally, manual background removal methods are implemented. Due to the large streams of data that need to be analyzed for the QPM, an automated background removal method is introduced that automatically discriminates the background from features of interest and characterizes and removes the background shape from all frames within a sequence of data. No user intervention is required and the performance rivals manual methods. The final step in processing data from a QPM is to ensure consistent phase unwrapping over an entire dataset. This is a previously undiscussed topic within the field of quantitative phase microscopy. The two-dimensional phase unwrapping methods result in reasonable phase estimates of the measured sample however there are often inconsistencies in local regions amongst sequential frames of data. This work introduces a new method, Smart Temporal unwrapping that minimizes temporal inconsistencies. The image processing methods presented in this work combine to allow phase data acquired using a QPM to quantify relative changes in biological samples. These processing steps effectively minimize errors due to system vibration, residual measurement aberration, and phase unwrapping inconsistencies.
| Identifer | oai:union.ndltd.org:arizona.edu/oai:arizona.openrepository.com:10150/311573 |
| Date | January 2013 |
| Creators | Goldstein, Goldie L. |
| Contributors | Wyant, James C., Wyant, James C., Creath, Katherine, Greivenkamp, John |
| Publisher | The University of Arizona. |
| Source Sets | University of Arizona |
| Language | en_US |
| Detected Language | English |
| Type | text, Electronic Dissertation |
| Rights | Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author. |
Page generated in 0.0011 seconds