文件夹修复

Question_41_50/README.md

@@ -5,7 +5,7 @@
 问题41至问题43是边缘检测方法中的一种——Canny 边缘检测法的理论介绍。
 
 1. 使用高斯滤波；
-2. 在x方向和y方向上使用 Sobel 滤波器，在此之上求出边缘的强度和边缘的梯度；
+2. 在$x$方向和$y$方向上使用Sobel滤波器，在此之上求出边缘的强度和边缘的梯度；
 3. 对梯度幅值进行非极大值抑制（Non-maximum suppression）来使边缘变得更细；
 4. 使用滞后阈值来对阈值进行处理。
 
@@ -212,9 +212,11 @@ x = - sin(t) / cos(t) * y + r / cos(t)
 
 ## Q.50. 闭运算（Closing Operation）
 
-Canny 边缘检测之后，进行闭处理吧（N=1）。
+Canny边缘检测之后，进行$N=1$的闭处理吧。
+
+闭运算，即先进行$N$次膨胀再进行$N$次腐蚀。
+
 
-闭运算，即先进行N次膨胀再进行N次腐蚀。
 
 闭运算能够将中断的像素连接起来。
 

Question_61_70/answers/answer_61.py

@@ -2,38 +2,54 @@
 import numpy as np
 import matplotlib.pyplot as plt
 
+# Connect 4
+def connect_4(img):
+    # get shape
+    H, W, C = img.shape
+
+    # prepare temporary image
+    tmp = np.zeros((H, W), dtype=np.int)
+
+    # binarize
+    tmp[img[..., 0] > 0] = 1
+
+    # prepare out image
+    out = np.zeros((H, W, 3), dtype=np.uint8)
+
+    # each pixel
+    for y in range(H):
+        for x in range(W):
+            if tmp[y, x] < 1:
+                continue
+
+            S = 0
+            S += (tmp[y,min(x+1,W-1)] - tmp[y,min(x+1,W-1)] * tmp[max(y-1,0),min(x+1,W-1)] * tmp[max(y-1,0),x])
+            S += (tmp[max(y-1,0),x] - tmp[max(y-1,0),x] * tmp[max(y-1,0),max(x-1,0)] * tmp[y,max(x-1,0)])
+            S += (tmp[y,max(x-1,0)] - tmp[y,max(x-1,0)] * tmp[min(y+1,H-1),max(x-1,0)] * tmp[min(y+1,H-1),x])
+            S += (tmp[min(y+1,H-1),x] - tmp[min(y+1,H-1),x] * tmp[min(y+1,H-1),min(x+1,W-1)] * tmp[y,min(x+1,W-1)])
+
+            if S == 0:
+                out[y,x] = [0, 0, 255]
+            elif S == 1:
+                out[y,x] = [0, 255, 0]
+            elif S == 2:
+                out[y,x] = [255, 0, 0]
+            elif S == 3:
+                out[y,x] = [255, 255, 0]
+            elif S == 4:
+                out[y,x] = [255, 0, 255]
+
+    out = out.astype(np.uint8)
+
+    return out
+
+
+
 # Read image
 img = cv2.imread("renketsu.png").astype(np.float32)
-H, W, C = img.shape
-
-tmp = np.zeros((H, W), dtype=np.int)
-tmp[img[..., 0]>0] = 1
-
-out = np.zeros((H, W, 3), dtype=np.uint8)
-
-for y in range(H):
-    for x in range(W):
-        if tmp[y, x] < 1:
-            continue
-
-        c = 0
-        c += (tmp[y,min(x+1,W-1)] - tmp[y,min(x+1,W-1)] * tmp[max(y-1,0),min(x+1,W-1)] * tmp[max(y-1,0),x])
-        c += (tmp[max(y-1,0),x] - tmp[max(y-1,0),x] * tmp[max(y-1,0),max(x-1,0)] * tmp[y,max(x-1,0)])
-        c += (tmp[y,max(x-1,0)] - tmp[y,max(x-1,0)] * tmp[min(y+1,H-1),max(x-1,0)] * tmp[min(y+1,H-1),x])
-        c += (tmp[min(y+1,H-1),x] - tmp[min(y+1,H-1),x] * tmp[min(y+1,H-1),min(x+1,W-1)] * tmp[y,min(x+1,W-1)])
-
-        if c == 0:
-            out[y,x] = [0, 0, 255]
-        elif c == 1:
-            out[y,x] = [0, 255, 0]
-        elif c == 2:
-            out[y,x] = [255, 0, 0]
-        elif c == 3:
-            out[y,x] = [255, 255, 0]
-        elif c == 4:
-            out[y,x] = [255, 0, 255]
-
-out = out.astype(np.uint8)
+
+# connect 4
+out = connect_4(img)
 
 # Save result
 cv2.imwrite("out.png", out)

Question_61_70/answers/answer_62.py

@@ -2,40 +2,57 @@
 import numpy as np
 import matplotlib.pyplot as plt
 
+# connect 8
+def connect_8(img):
+    # get shape
+    H, W, C = img.shape
+
+    # prepare temporary
+    _tmp = np.zeros((H, W), dtype=np.int)
+
+    # get binarize
+    _tmp[img[..., 0] > 0] = 1
+
+    # inverse for connect 8
+    tmp = 1 - _tmp
+
+    # prepare image
+    out = np.zeros((H, W, 3), dtype=np.uint8)
+
+    # each pixel
+    for y in range(H):
+        for x in range(W):
+            if _tmp[y, x] < 1:
+                continue
+
+            S = 0
+            S += (tmp[y,min(x+1,W-1)] - tmp[y,min(x+1,W-1)] * tmp[max(y-1,0),min(x+1,W-1)] * tmp[max(y-1,0),x])
+            S += (tmp[max(y-1,0),x] - tmp[max(y-1,0),x] * tmp[max(y-1,0),max(x-1,0)] * tmp[y,max(x-1,0)])
+            S += (tmp[y,max(x-1,0)] - tmp[y,max(x-1,0)] * tmp[min(y+1,H-1),max(x-1,0)] * tmp[min(y+1,H-1),x])
+            S += (tmp[min(y+1,H-1),x] - tmp[min(y+1,H-1),x] * tmp[min(y+1,H-1),min(x+1,W-1)] * tmp[y,min(x+1,W-1)])
+
+            if S == 0:
+                out[y,x] = [0, 0, 255]
+            elif S == 1:
+                out[y,x] = [0, 255, 0]
+            elif S == 2:
+                out[y,x] = [255, 0, 0]
+            elif S == 3:
+                out[y,x] = [255, 255, 0]
+            elif S == 4:
+                out[y,x] = [255, 0, 255]
+
+    out = out.astype(np.uint8)
+
+    return out
+
+
 # Read image
 img = cv2.imread("renketsu.png").astype(np.float32)
-H, W, C = img.shape
-
-_tmp = np.zeros((H, W), dtype=np.int)
-_tmp[img[..., 0]>0] = 1
-
-tmp = 1 - _tmp
-
-out = np.zeros((H, W, 3), dtype=np.uint8)
-
-for y in range(H):
-    for x in range(W):
-        if _tmp[y, x] < 1:
-            continue
-
-        c = 0
-        c += (tmp[y,min(x+1,W-1)] - tmp[y,min(x+1,W-1)] * tmp[max(y-1,0),min(x+1,W-1)] * tmp[max(y-1,0),x])
-        c += (tmp[max(y-1,0),x] - tmp[max(y-1,0),x] * tmp[max(y-1,0),max(x-1,0)] * tmp[y,max(x-1,0)])
-        c += (tmp[y,max(x-1,0)] - tmp[y,max(x-1,0)] * tmp[min(y+1,H-1),max(x-1,0)] * tmp[min(y+1,H-1),x])
-        c += (tmp[min(y+1,H-1),x] - tmp[min(y+1,H-1),x] * tmp[min(y+1,H-1),min(x+1,W-1)] * tmp[y,min(x+1,W-1)])
-
-        if c == 0:
-            out[y,x] = [0, 0, 255]
-        elif c == 1:
-            out[y,x] = [0, 255, 0]
-        elif c == 2:
-            out[y,x] = [255, 0, 0]
-        elif c == 3:
-            out[y,x] = [255, 255, 0]
-        elif c == 4:
-            out[y,x] = [255, 0, 255]
-
-out = out.astype(np.uint8)
+
+# connect 8
+out = connect_8(img)
+
 
 # Save result
 cv2.imwrite("out.png", out)

Question_61_70/answers/answer_63.py

@@ -2,46 +2,62 @@
 import numpy as np
 import matplotlib.pyplot as plt
 
-# Read image
-img = cv2.imread("gazo.png").astype(np.float32)
-H, W, C = img.shape
-
-out = np.zeros((H, W), dtype=np.int)
-out[img[..., 0]>0] = 1
-
-count = 1
-while count > 0:
-    count = 0
-    tmp = out.copy()
-    for y in range(H):
-        for x in range(W):
-            if out[y, x] < 1:
-                continue
-
-            judge = 0
-
-            ## condition 1
-            if (tmp[y,min(x+1,W-1)] + tmp[max(y-1,0), x] + tmp[y,max(x-1,0)] + tmp[min(y+1,H-1),x]) < 4:
-                judge += 1
+
+# thining algorythm
+def thining(img):
+    # get shape
+    H, W, C = img.shape
+
+    # prepare out image
+    out = np.zeros((H, W), dtype=np.int)
+    out[img[..., 0] > 0] = 1
+
+    count = 1
+    while count > 0:
+        count = 0
+        tmp = out.copy()
+        # each pixel ( rasta scan )
+        for y in range(H):
+            for x in range(W):
+                # skip black pixel
+                if out[y, x] < 1:
+                    continue
+
+                # count satisfied conditions
+                judge = 0
 
-            ## condition 2
-            c = 0
-            c += (out[y,min(x+1,W-1)] - out[y,min(x+1,W-1)]*out[max(y-1,0),min(x+1,W-1)]*out[max(y-1,0),x])
-            c += (out[max(y-1,0),x] - out[max(y-1,0),x]*out[max(y-1,0),max(x-1,0)]*out[y,max(x-1,0)])
-            c += (out[y,max(x-1,0)] - out[y,max(x-1,0)]*out[min(y+1,H-1),max(x-1,0)]*out[min(y+1,H-1),x])
-            c += (out[min(y+1,H-1),x] - out[min(y+1,H-1),x]*out[min(y+1,H-1),min(x+1,W-1)]*out[y,min(x+1,W-1)])
-            if c == 1:
-                judge += 1
+                ## condition 1
+                if (tmp[y, min(x+1, W-1)] + tmp[max(y-1, 0), x] + tmp[y, max(x-1, 0)] + tmp[min(y+1, H-1), x]) < 4:
+                    judge += 1
+
+                ## condition 2
+                c = 0
+                c += (tmp[y,min(x+1, W-1)] - tmp[y, min(x+1, W-1)] * tmp[max(y-1, 0),min(x+1, W-1)] * tmp[max(y-1, 0), x])
+                c += (tmp[max(y-1,0), x] - tmp[max(y-1,0), x] * tmp[max(y-1, 0), max(x-1, 0)] * tmp[y, max(x-1, 0)])
+                c += (tmp[y, max(x-1, 0)] - tmp[y,max(x-1, 0)] * tmp[min(y+1, H-1), max(x-1, 0)] * tmp[min(y+1, H-1), x])
+                c += (tmp[min(y+1, H-1), x] - tmp[min(y+1, H-1), x] * tmp[min(y+1, H-1), min(x+1, W-1)] * tmp[y, min(x+1, W-1)])
+                if c == 1:
+                    judge += 1
+
+                ##x condition 3
+                if np.sum(tmp[max(y-1, 0) : min(y+2, H), max(x-1, 0) : min(x+2, W)]) >= 4:
+                    judge += 1
 
-            ##x condition 3
-            if np.sum(out[max(y-1,0):min(y+2,H), max(x-1,0):min(x+2,W)]) >= 4:
-                judge += 1
-
-            if judge == 3:
-                out[y,x] = 0
-                count += 1
-
-out = out.astype(np.uint8) * 255
+                # if all conditions are satisfied
+                if judge == 3:
+                    out[y, x] = 0
+                    count += 1
+
+    out = out.astype(np.uint8) * 255
+
+    return out
+
+
+# Read image
+img = cv2.imread("gazo.png").astype(np.float32)
+
+# thining
+out = thining(img)
 
 # Save result
 cv2.imwrite("out.png", out)