Well, with fluorescent imaging, you barely need machine learning as nothing else in the image is visible besides the labelled organisms.
-- ie. for identifying normal bacterial colonies, I have to use a dozen or so filters, to correct for background illumination, do complex adaptive thresholding etc etc. For identifying fluorescent bacterial colonies, I just threshold all pixels that are 3x the standard deviation of the image. For bacteria, you could then just do a watershed, and filter on the size of the detected blobs.
As far as finding filters goes, search for "dichroic mirror" on ebay, and select the filter you need from the spectral info. for fluorescent imaging, make sure the excitation and emission bandpass ranges are seperated by at least 15nm to limit bleedthrough. If you are not doing multispectral imaging (eg GFP, YFP and RFP together), then excite at the blue end of the emission spectra, and use a longpass filter at the emission maxima of the fluorescent protein you are imaging. You'll need a dichroic mirror to separate the excitation and emission light too. It's cutoff should be in between the bandpass ranges.
You can look at the spectra and plan using this web tool:
https://www.lifetechnologies.com/de/de/home/life-science/cell-analysis/labeling-chemistry/fluorescence-spectraviewer.html
-Dave
On Saturday, June 28, 2014 7:22:17 PM UTC+2, Michael Shamberger wrote:
On Saturday, June 28, 2014 7:22:17 PM UTC+2, Michael Shamberger wrote:
They are updating the microscope design this summer. So far they have tracked a nematode.>> found the openlabtools data a few months ago
and ordered what should be an appropriate tube lens for monochromatic
use.What did you end up buying?>> Why do you want to start with fluorescense? Is there anything youactually want to see using this?I want to see if I can track/image live bacteria with a DIY setup costing less the 1K. Then I want to try various machine learning techniques on the images to see if I can reproduce some of the latest research in this area.
-- You received this message because you are subscribed to the Google Groups DIYbio group. To post to this group, send email to diybio@googlegroups.com. To unsubscribe from this group, send email to diybio+unsubscribe@googlegroups.com. For more options, visit this group at https://groups.google.com/d/forum/diybio?hl=en
Learn more at www.diybio.org
---
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diybio+unsubscribe@googlegroups.com.
To post to this group, send email to diybio@googlegroups.com.
Visit this group at http://groups.google.com/group/diybio.
To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/e8bfe39d-3acb-45bb-98ee-2acd4018e02d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
0 comments:
Post a Comment