Five image processing algorithms are proposed to measure the average orientation, eccentricity and size of cells in images of biological tissue. These properties, which can be embodied by an elliptical 'composite cell', are crucial for biomechanical tissue models. To automatically determine these properties is challenging due to the diverse nature of the image data, with tremendous and unpredictable variability in illumination, cell pigmentation, cell shape and cell boundary visibility. One proposed algorithm estimates the composite cell properties directly from the input tissue image, while four others estimate the properties from frequency domain data. The accuracy and stability of the algorithms are quantitatively compared through application to a wide variety of real images. Based on these results, the best algorithm is selected.
| Identifer | oai:union.ndltd.org:WATERLOO/oai:uwspace.uwaterloo.ca:10012/905 |
| Date | January 2005 |
| Creators | Iles, Peter |
| Publisher | University of Waterloo |
| Source Sets | University of Waterloo Electronic Theses Repository |
| Language | English |
| Detected Language | English |
| Type | Thesis or Dissertation |
| Format | application/pdf, 2303770 bytes, application/pdf |
| Rights | Copyright: 2005, Iles, Peter. All rights reserved. |
Page generated in 0.0021 seconds