I am a novice in image processing and I hope I am not asking too much at once
I have lambda scans consisting of 32 channels (411-695 nm). They are converted to stacks in tif format to use with an ImageJ plugin. (spectral phasor)
I want to find a standardized and automatic way of setting a threshold to use with the plugin. At the thresholding step, the plugin sums the pixel values along all channels and counts the pixels with larger values than the given threshold to filter out the background (as far as I understand from the script)
Image-DAPI-BODIPY-Texas+red.tif (12.3 MB)
I start with this example image from the plugin website. In the example they used 8000 as the threshold. So I am trying to obtain a similar value by using an auto-thresholding method. I have tried Image > Adjust > Auto threshold and Threshold or starting with showing the histogram to see where the threshold could be set.
- My first question is: If I show the histogram for the stack by including all 128 channels it looks like this:
Histogram (216.4 KB)
I don’t understand the stack histogram very well. Is it a cumulative histogram?
- I guess in order to obtain a similar threshold to what was recommended, I should use the sum of all slices. But if I do Z project and choose “Sum slices”, the resulting image is 32-bit. To be able to use “Auto threshold” I have to convert it to 16-bit. In that case the range of the pixel values changes, thus the location of the threshold on the histogram does too. How can I apply auto-thresholding on the sum image?
- This is one of the images I am actually working with.
The histogram of the stack looks like this:
If I sum all slices the same way, the histogram shows the smallest pixel value to be 3236.
It doesn’t seem possible to me that there are no coinciding 0-value pixels among the slices. What could be the reason for the values to start from 3000s?
I appreciate any suggestions and help. Thanks in advance