Histogram Equalization¶
This is a method used to normalize the distribution of signal intensity values within an image. If the image has low contrast it will make it easier to threshold.
hist_equalization(img, device, debug=None)
returns device, normalized image
- Parameters:
- img - the original 2 dimensional grayscale image for analysis.
- device - Counter for image processing steps
- debug- None, "print", or "plot". Print = save to file, Plot = print to screen. Default = None
- Context:
- Used to normalize the distribution of a signal intensity within an image.
Grayscale image
from plantcv import plantcv as pcv
# Examine signal distribution within an image
# prints out an image histogram of signal within image
device, he_img = pcv.HistEqualization(img, device, debug="print")
Normalized image