完成频域的锐化

basic/urls.py

@@ -19,4 +19,5 @@ urlpatterns = [
     path('ed', views.edge_detection),
     path('lcd', views.line_change_detection),
     path('smooth', views.smooth),
+    path('sharpen', views.sharpen),
 ]

basic/views.py

@@ -609,6 +609,18 @@ def ideal_low_filter(img, D0):
     return filter_img
 
 
+def ideal_high_pass_filter(img, D0):
+    h, w = img.shape[:2]
+    filter_img = np.zeros((h, w))
+    u = np.fix(h / 2)
+    v = np.fix(w / 2)
+    for i in range(h):
+        for j in range(w):
+            d = np.sqrt((i - u) ** 2 + (j - v) ** 2)
+            filter_img[i, j] = 0 if d < D0 else 1
+    return filter_img
+
+
 def butterworth_low_filter(img, D0, rank):
     """
         生成一个Butterworth低通滤波器（并返回）
@@ -624,6 +636,21 @@ def butterworth_low_filter(img, D0, rank):
     return filter_img
 
 
+def butterworth_high_pass_filter(img, D0, rank):
+    """
+        生成一个Butterworth低通滤波器（并返回）
+    """
+    h, w = img.shape[:2]
+    filter_img = np.zeros((h, w))
+    u = np.fix(h / 2)
+    v = np.fix(w / 2)
+    for i in range(h):
+        for j in range(w):
+            d = np.sqrt((i - u) ** 2 + (j - v) ** 2)
+            filter_img[i, j] = 1 / (1 + ((D0 / d) ** (2 * rank)))
+    return filter_img
+
+
 def gauss_low_pass_filter(img, d0):
     """
         生成一个指数低通滤波器（并返回）
@@ -639,7 +666,22 @@ def gauss_low_pass_filter(img, d0):
     return filter_img
 
 
-def filter_use(img, filter):
+def gauss_high_pass_filter(img, d0):
+    """
+        生成一个指数低通滤波器（并返回）
+    """
+    h, w = img.shape[:2]
+    filter_img = np.zeros((h, w))
+    u = np.fix(h / 2)
+    v = np.fix(w / 2)
+    for i in range(h):
+        for j in range(w):
+            d = np.sqrt((i - u) ** 2 + (j - v) ** 2)
+            filter_img[i, j] = 1 - np.exp(-0.5 * (d ** 2) / (d0 ** 2))
+    return filter_img
+
+
+def filter_use_smooth(img, filter):
     """
     将图像img与滤波器filter结合，生成对应的滤波图像
     """
@@ -654,6 +696,18 @@ def filter_use(img, filter):
     return f_origin
 
 
+def filter_use_sharpen(img, filter):
+    f = np.fft.fft2(img)
+    f_center = np.fft.fftshift(f)
+    # 应用滤波器进行反变换
+    S = np.multiply(f_center, filter)  # 频率相乘——l(u,v)*H(u,v)
+    f_origin = np.fft.ifftshift(S)  # 将低频移动到原来的位置
+    f_origin = np.fft.ifft2(f_origin)  # 使用ifft2进行傅里叶的逆变换
+    f_origin = np.abs(f_origin)  # 设置区间
+    f_origin = f_origin / np.max(f_origin.all())
+    return f_origin
+
+
 @csrf_exempt
 def smooth(request):
     if request.method == 'POST':
@@ -662,18 +716,46 @@ def smooth(request):
         img = cv2.imread(PREFIX + image, 0)
         # frequency domain
         fd = para['fd']
-        smooth_type = para['type']
+        filter = para['filter']
         if fd:
-            if smooth_type == 'ideal_low_pass_filter':
+            if filter == 'ideal_low_pass_filter':
+                d0 = para.get('d0', 20)
+                img = filter_use_smooth(img, ideal_low_filter(img, d0))
+            elif filter == 'butterworth_low_pass_filter':
+                d0 = para.get('d0', 20)
+                rank = para.get('rank', 2)
+                img = filter_use_smooth(img, butterworth_low_filter(img, d0, rank))
+            elif filter == 'gauss_low_pass_filter':
                 d0 = para.get('d0', 20)
-                img = filter_use(img, ideal_low_filter(img, d0))
-            elif smooth_type == 'butterworth_low_pass_filter':
+                img = filter_use_smooth(img, gauss_low_pass_filter(img, d0))
+        else:
+            return 0
+        filename = getImageName() + DEFAULT_FORMAT
+        cv2.imwrite(PREFIX + filename, img)
+        return HttpResponse(filename)
+    return HttpResponse('请使用POST方法')
+
+
+@csrf_exempt
+def sharpen(request):
+    if request.method == 'POST':
+        para = json.loads(request.body)
+        image = para['img']
+        img = cv2.imread(PREFIX + image, 0)
+        # frequency domain
+        fd = para['fd']
+        filter = para['filter']
+        if fd:
+            if filter == 'ideal_high_pass_filter':
+                d0 = para.get('d0', 40)
+                img = filter_use_sharpen(img, ideal_high_pass_filter(img, d0))
+            elif filter == 'butterworth_high_pass_filter':
                 d0 = para.get('d0', 20)
                 rank = para.get('rank', 2)
-                img = filter_use(img, butterworth_low_filter(img, d0, rank))
-            elif smooth_type == 'gauss_low_pass_filter':
+                img = filter_use_sharpen(img, butterworth_high_pass_filter(img, d0, rank))
+            elif filter == 'gauss_high_pass_filter':
                 d0 = para.get('d0', 20)
-                img = filter_use(img, gauss_low_pass_filter(img, d0))
+                img = filter_use_sharpen(img, gauss_high_pass_filter(img, d0))
         else:
             return 0
         filename = getImageName() + DEFAULT_FORMAT