Dithering (D4, D8)

Output 4 gray values and 8 gray values

Author: sunnu06

.DS_Store

@@ -0,0 +1,50 @@
+# Implement Dithering extend to n=4 gray values
+from PIL import Image
+import numpy as np
+
+# Input a grayscale image I
+I = Image.open("my_cat.png").convert('L')
+width, height = I.size
+
+# Step1 : n output and m range Declaration
+n = 4
+m = int(255/(n-1))
+
+# Step2 : Define and Calculate Q
+column = height
+row = width
+# Define Q (2D Array)
+Q = [[0 for _ in range(row)] for _ in range(column)]
+# Calculate Q(i,j) value
+for j in range(I.size[1]):      # j = column
+    for i in range(I.size[0]):  # i = row
+        Q[j][i] = int(I.getpixel((i, j))/m)  # Q(i,j) = [I(i,j)/85]
+
+# Step3 :Define dithering matrix(D1) and generate dithering array(D)
+# Define Dithering Matrix (2X2)
+D1 = np.array([[0, 56],
+              [84, 28]])
+
+# Generate Dithering Array D of image size by repeating D1
+D = np.tile(D1, (int(width/2), int(height/2)))
+
+# Step4 : Threshold image I
+for j in range(I.size[1]):      # j = column
+    for i in range(I.size[0]):  # i = row
+
+        ori_pixel = I.getpixel((i, j))
+
+        if ori_pixel - m * Q[j][i] > D[j][i]:
+            I.putpixel((i, j), m * (Q[j][i] + 1))
+        else:
+            I.putpixel((i, j), m * (Q[j][i] + 0))
+
+# Save new image I'
+I.save("D4.png")
+
+# Step5 : Display images I and I'(dithering_img)
+I = Image.open("my_cat.png").convert('L')
+I.show()
+dithering_img = Image.open("D4.png")
+dithering_img.show()
+

01.Dithering/D4.png

@@ -0,0 +1,50 @@
+# Implement Dithering extend to n=8 gray values
+from PIL import Image
+import numpy as np
+
+# Input a grayscale image I
+I = Image.open("my_cat.png").convert('L')
+width, height = I.size
+
+# Step1 : n output and m range Declaration
+n = 8
+m = int(255/(n-1))
+
+# Step2 : Define and Calculate Q
+column = height
+row = width
+# Define Q (2D Array)
+Q = [[0 for _ in range(row)] for _ in range(column)]
+# Calculate Q(i,j) value
+for j in range(I.size[1]):      # j = column
+    for i in range(I.size[0]):  # i = row
+        Q[j][i] = int(I.getpixel((i, j))/m)  # Q(i,j) = [I(i,j)/85]
+
+# Step3 :Define dithering matrix(D1) and generate dithering array(D)
+# Define Dithering Matrix (2X2)
+D1 = np.array([[0, 24],
+              [36, 12]])
+
+# Generate Dithering Array D of image size by repeating D1
+D = np.tile(D1, (int(width/2), int(height/2)))
+
+# Step4 : Threshold image I
+for j in range(I.size[1]):      # j = column
+    for i in range(I.size[0]):  # i = row
+
+        ori_pixel = I.getpixel((i, j))
+
+        if ori_pixel - m * Q[j][i] > D[j][i]:
+            I.putpixel((i, j), m * (Q[j][i] + 1))
+        else:
+            I.putpixel((i, j), m * (Q[j][i] + 0))
+
+# Save new image I'
+I.save("D8.png")
+
+# Step5 : Display images I and I'(dithering_img)
+I = Image.open("my_cat.png").convert('L')
+I.show()
+dithering_img = Image.open("D8.png")
+dithering_img.show()
+