Testing scanners

Testing stability of a scanner

Place a fixture or similar object in a scanner and start an experiment.

Note that you should not have any plates or things that change over time in the scanner

Afterwards, in ipython from the directory where the images are:

from matplotlib import pyplot as plt
from glob import glob
import numpy as np
counts = []

files = sorted(glob("*.tiff"))
for file in files:
    im = plt.imread(file)
    count = np.bincount(im.ravel())
    counts.append(count)

counts = np.array(counts)
means = counts.mean(axis=0)
std = counts.std(axis=0)

plt.ion()

plt.bar(np.arange(means.size), height=means, width=1.0, yerr=std)
plt.title('Scan-o-Matic, default process, 10 minutes between scans')  # Change this to what makes sense for you
plt.xlabel('Pixel intensity')
plt.xlim(0, 256)
plt.ylabel('Frequency in image & Standard dev')

plt.show()

If you use the same standard motif, the same spectrum should be evident between tests and within the spectrum the noise/error should be very small.