Closed scottclowe closed 9 years ago
Might be easier to compute with mahotas than with skimage.
http://mahotas.readthedocs.org/en/latest/api.html?highlight=haralick#mahotas.features.haralick
This should subclass HighLevelFeatures
, and be computed with mahotas.features.haralick
using the return_mean_ptp
flag to take the angular mean.
Extract Haralick texture features from the image. Apparently, these are popular in biological CV.
http://murphylab.web.cmu.edu/publications/boland/boland_node26.html http://scikit-image.org/docs/dev/auto_examples/plot_glcm.html