Mahotas – Computing Linear Binary Patterns

In this article we will see how we can get the linear binary patterns of image in mahotas. Local binary patterns is a type of visual descriptor used for classification in computer vision. LBP is the particular case of the Texture Spectrum model proposed in 1990. LBP was first described in 1994. For this we are going to use the fluorescent microscopy image from a nuclear segmentation benchmark. We can get the image with the help of command given below

mahotas.demos.nuclear_image()

Below is the nuclear_image 

In order to do this we will use mahotas.features.lbp method 

Syntax : mahotas.features.lbp(image, radius, points)
Argument : It takes image object and two integers as argument
Return : It returns 1-D numpy ndarray i.e histogram feature 

Note : The input of the this should  be the filtered image or loaded as grey
In order to filter the image we will take the image object which is numpy.ndarray and filter it with the help of indexing, below is the command to do this 

image = image[:, :, 0]

Example 1 :

Python3

# importing various libraries
import mahotas
import mahotas.demos
import mahotas as mh
import numpy as np
from pylab import imshow, show
import matplotlib.pyplot as plt

# loading nuclear image
nuclear = mahotas.demos.nuclear_image()

# filtering image
nuclear = nuclear[:, :, 0]

# adding gaussian filter
nuclear = mahotas.gaussian_filter(nuclear, 4)

# setting threshold
threshed = (nuclear > nuclear.mean())

# making is labelled image
labeled, n = mahotas.label(threshed)

# showing image
print("Labelled Image")
imshow(labelled)
show()

# Computing Linear Binary Patterns
value = mahotas.features.lbp(labelled, 200, 5)

# showing histograph
plt.hist(value)

Output :

Example 2 :

Python3

# importing required libraries
import numpy as np
import mahotas
from pylab import imshow, show
import matplotlib.pyplot as plt

# loading image
img = mahotas.imread('dog_image.png')

# filtering the image
img = img[:, :, 0]

# setting gaussian filter
gaussian = mahotas.gaussian_filter(img, 15)

# setting threshold value
gaussian = (gaussian > gaussian.mean())

# making is labelled image
labeled, n = mahotas.label(gaussian)

# showing image
print("Labelled Image")
imshow(labelled)
show()


# Computing Linear Binary Patterns
value = mahotas.features.lbp(labelled, 200, 5, ignore_zeros = False)


# showing histograph
plt.hist(value)

Output :