spatial filters

hi-pass.py

@@ -0,0 +1,57 @@
+def clamp_pixel_value(value):
+  if value < 0:
+    return 0
+  elif value > 255:
+    return 255
+  else:
+    return value
+
+def LaplacianFilter(img, mask=None):
+  assert (img is not None), "The image is not loaded"
+
+  width, height = img.shape
+  padded_width = (width//3) * 3
+  padded_height = (height//3) * 3
+  new_img = img.copy()
+
+  assert (new_img is not None), "Memory not enough to allocate storage for new image"
+
+  laplacian_mask = None
+
+  if mask == "variant":
+    laplacian_mask = [[1, 1, 1], [1, -8, 1], [1, 1, 1]]
+  elif mask == "sobelx":
+    laplacian_mask = [[-1, -2, -1], [0, 0, 0], [1, 2, 1]]
+  elif mask == "sobely":
+    laplacian_mask = [[-1, 0, 1], [-2, 0, 2], [-1, 0, 1]]
+  else:
+    laplacian_mask = [[0, 1, 0], [1, -4, 1], [0, 1, 0]]
+
+  for i in range(1, padded_width-1):
+    for j in range(1, padded_height-1):
+      sum = 0
+
+      for k in range(i-1, i+2):
+        for w in range(j-1, j+2):
+            sum += img[k][w] * laplacian_mask[k-i][w-j] 
+
+      new_img[i][j] = clamp_pixel_value(sum)
+
+  return new_img
+
+def LaplacianEnhance(img, filter):
+  assert (img is not None), "The image is not loaded"
+  assert (filter is not None), "The filter is not loaded"
+
+  img_width, img_height = img.shape
+  filter_width, filter_height = filter.shape
+
+  assert (img_width == filter_width and img_height == filter_height, "Image and filter sizes are not matching")
+
+  new_img = img.copy()
+
+  for i in range(img_width):
+    for j in range(img_height):
+      new_img[i][j] = img[i][j] - filter[i][j]
+
+  return new_img

low-pass.py

@@ -0,0 +1,93 @@
+def MinFilter(img, nbhd_size):
+  assert (nbhd_size % 2 != 0), "The neighbourhood size must not be even"
+  assert (img is not None), "The image is not loaded"
+
+  width, height = img.shape
+  padded_width = (width//nbhd_size) * nbhd_size
+  padded_height = (height//nbhd_size) * nbhd_size
+  new_img = img.copy()
+
+  assert (new_img is not None), "Memory not enough to allocate storage for new image"
+
+  for i in range(((nbhd_size-1)//2), padded_width-((nbhd_size-1)//2)):
+    for j in range(((nbhd_size-1)//2), padded_height-((nbhd_size-1)//2)):
+      min = 255
+
+      for k in range(i-((nbhd_size-1)//2), i+((nbhd_size-1)//2)+1):
+        for w in range(j-((nbhd_size-1)//2), j+((nbhd_size-1)//2)+1):
+          if img[k][w] < min:
+            min = img[k][w]
+
+      new_img[i][j] = min
+
+  return new_img
+
+def MaxFilter(img, nbhd_size):
+  assert (nbhd_size % 2 != 0), "The neighbourhood size must not be even"
+  assert (img is not None), "The image is not loaded"
+
+  width, height = img.shape
+  padded_width = (width//nbhd_size) * nbhd_size
+  padded_height = (height//nbhd_size) * nbhd_size
+  new_img = img.copy()
+
+  assert (new_img is not None), "Memory not enough to allocate storage for new image"
+
+  for i in range(((nbhd_size-1)//2), padded_width-((nbhd_size-1)//2)):
+    for j in range(((nbhd_size-1)//2), padded_height-((nbhd_size-1)//2)):
+      max = 0
+
+      for k in range(i-((nbhd_size-1)//2), i+((nbhd_size-1)//2)+1):
+        for w in range(j-((nbhd_size-1)//2), j+((nbhd_size-1)//2)+1):
+          if img[k][w] > max:
+            max = img[k][w]
+
+      new_img[i][j] = max
+
+  return new_img
+
+def MeanFilter(img, nbhd_size):
+  assert (nbhd_size % 2 != 0), "The neighbourhood size must not be even"
+  assert (img is not None), "The image is not loaded"
+
+  width, height = img.shape
+  padded_width = (width//nbhd_size) * nbhd_size
+  padded_height = (height//nbhd_size) * nbhd_size
+  new_img = img.copy()
+
+  assert (new_img is not None), "Memory not enough to allocate storage for new image"
+
+  for i in range(((nbhd_size-1)//2), padded_width-((nbhd_size-1)//2)):
+    for j in range(((nbhd_size-1)//2), padded_height-((nbhd_size-1)//2)):
+      sum = 0
+
+      for k in range(i-((nbhd_size-1)//2), i+((nbhd_size-1)//2)+1):
+        for w in range(j-((nbhd_size-1)//2), j+((nbhd_size-1)//2)+1):
+            sum += img[k][w]
+
+      new_img[i][j] = sum/(nbhd_size**2 - 1)
+
+  return new_img
+
+def MedianFilter(img, nbhd_size):
+  assert (nbhd_size % 2 != 0), "The neighbourhood size must not be even"
+  assert (img is not None), "The image is not loaded"
+
+  width, height = img.shape
+  padded_width = (width//nbhd_size) * nbhd_size
+  padded_height = (height//nbhd_size) * nbhd_size
+  new_img = img.copy()
+
+  assert (new_img is not None), "Memory not enough to allocate storage for new image"
+
+  for i in range(((nbhd_size-1)//2), padded_width-((nbhd_size-1)//2)):
+    for j in range(((nbhd_size-1)//2), padded_height-((nbhd_size-1)//2)):
+      vals = []
+
+      for k in range(i-((nbhd_size-1)//2), i+((nbhd_size-1)//2)+1):
+        for w in range(j-((nbhd_size-1)//2), j+((nbhd_size-1)//2)+1):
+            vals.append(img[k][w])
+
+      new_img[i][j] = vals[len(vals)//2]
+
+  return new_img