Lab 3

# -*- coding: utf-8 -*-
"""
Created on Wed Mar 15 18:50:19 2023

@author: joshu
"""

#%% Lab 3: Image binarization. Dithering.

#%% Libraries

import numpy as np
import matplotlib.pyplot as plt

from skimage import data 
from scipy import signal
from skimage.filters import threshold_otsu

#%% 3.1 Adaptive Thresholding

im = data.retina()

# We have to calculate the Luminance to be able to display the gray level image.

L = 0.299*np.double(im[:,:,0]) + 0.587*np.double(im[:,:,1]) + 0.114*np.double(im[:,:,2])


plt.figure('3.1')
plt.suptitle('3.1 Adaptive Thresholding')

plt.subplot(2, 2, 1)
plt.title('Original')
plt.imshow(L, cmap = 'gray')

# Global Thresholding

# Selecting the optimal threshold using the following command

th = np.uint8(threshold_otsu(L))/255
L_gth = np.uint8(L>=th)

plt.subplot(2,2,2)
plt.title('Global Thresholding')
plt.imshow(L_gth, cmap = 'gray')

# Median Thresholding


L_mth = signal.medfilt(L, kernel_size=3)

L_mf = L>=L_mth

plt.subplot(2,2,3)
plt.title('Median Thresholding')
plt.imshow(L_mf, cmap = 'gray')

# Order Filter

domain = np.identity(3)

L_oth = signal.order_filter(L, domain, 2)
L_of = L>=L_oth

plt.subplot(2,2,4)
plt.title('Median Thresholding: Diff domain')
plt.imshow(L_of, cmap = 'gray')

plt.tight_layout()