how to find glcm of an image in OpenCV or numpy?

Question

I have to find texture based features of an image like correlation, energy, homogeneity, and contrast in Python. I am using python-opencv to do other operations, since python-opencv returns image as an numpy array. But I didn't find any good resource to compute GLCM using either opencv or numpy. Also I have to extract feature for d=1 and angle=[0,45,135,90] and then use this feature in svm.

Answer 1

You have to install the following libraries scipy , numpy and skimage .

skimage has a module called skimage.feature which includes skimage.feature.greycomatrix and skimage.feature.greycoprops . Using these we can calculate the greycoimage for ndarray uint8 and also the greco proporties instructions