- Data Collection:
- Trial 1: We tried using the Classic Watershed options under the Morphological segmentation plugin. I did not turn out anywhere near what we thought it would and one of the reasons why we thought it didn’t work was because we converted the image (already in RGB) to CIELAB which specify lighting conditions. In other words, it turned the image to grayscale based on the lighting of the color stains.
- Morphological Segmentation is an ImageJ/Fiji plugin that combines morphological operations, such as extended minima and morphological gradient, with watershed flooding algorithms to segment grayscale images of any type.
- MorphoLIBJ (Morphological segmentation)
- Trial 2: Worked a lot better, however, we most likely will not be able to use this image. What we did different to yield better results was simply converting the image to basic grayscale. Comparing to the first image, the xylem region is not nearly as intensified.
2. Summarize Accomplishments and briefly explain next step:
- Overall, we were not able to successfully use the morphological segmentation to accomplish our goal. Although it can potentially help, throughout our trials, it did not really work.
- Mitch was helping me research and understand how to use the plugin. He watched tutorial videos, read online, and preformed trials as well.
