Complete 3

3/2.py

@@ -0,0 +1,60 @@
+import cv2 as cv
+import numpy as np
+from math import *  # type: ignore
+import random
+import matplotlib.pyplot as plt
+
+plt.rcParams["font.sans-serif"] = ["SimSun"]
+plt.rcParams["axes.unicode_minus"] = False
+
+
+def addGaussianNoise(src, means, sigma):
+    NoiseImg = src / src.max()
+    rows = NoiseImg.shape[0]
+    cols = NoiseImg.shape[1]
+    for i in range(rows):
+        for j in range(cols):
+            NoiseImg[i, j] = NoiseImg[i, j] + random.gauss(means, sigma)
+            if NoiseImg[i, j] < 0:
+                NoiseImg[i, j] = 0
+            elif NoiseImg[i, j] > 1:
+                NoiseImg[i, j] = 1
+    return NoiseImg
+
+
+if __name__ == "__main__":
+    img0 = cv.imread(r"img\peppers.bmp", cv.IMREAD_GRAYSCALE)
+    img = addGaussianNoise(img0, 0, 0.1)
+    f = np.fft.fft2(img)
+    fshift = np.fft.fftshift(f)
+    magnitude_spectrum0 = 20 * np.log(1 + np.abs(fshift))
+    plt.figure(figsize=(10, 5))
+    plt.subplot(141)
+    plt.imshow(img, cmap="gray")
+    plt.title("噪声图像")
+    plt.axis("off")
+    plt.subplot(142)
+    plt.imshow(magnitude_spectrum0, cmap="gray")
+    plt.title("噪声图像幅值谱")
+    plt.axis("off")
+    r = 50
+    m, n = fshift.shape
+    H = np.zeros((m, n))
+    for i in range(m):
+        for j in range(n):
+            d = sqrt((i - m / 2) ** 2 + (j - n / 2) ** 2)
+            if d < r:
+                H[i, j] = 1
+    G = H * fshift
+    magnitude_spectrum1 = 20 * np.log(1 + np.abs(G))
+    f1 = np.fft.ifftshift(G)
+    img1 = np.abs(np.fft.ifft2(f1))
+    plt.subplot(143)
+    plt.imshow(magnitude_spectrum1, cmap="gray")
+    plt.title("ILPF滤波后幅值谱")
+    plt.axis("off")
+    plt.subplot(144)
+    plt.imshow(img1, cmap="gray")
+    plt.title("ILPF滤波后重构图像")
+    plt.axis("off")
+    plt.show()

3/3.py

@@ -0,0 +1,52 @@
+# 编写程序实现巴特沃斯低通滤波，H(u,v)=1/(1+[D(u,v)/D0]^2n)，实验图8是阶数n=4时的滤波效果图，改变n值，查看并分析阶数n对滤波器的影响
+
+import cv2 as cv
+import numpy as np
+from math import *  # type: ignore
+import random
+import matplotlib.pyplot as plt
+
+plt.rcParams["font.sans-serif"] = ["SimSun"]
+plt.rcParams["axes.unicode_minus"] = False
+
+
+if __name__ == "__main__":
+    img = cv.imread(r"img\alphabet.jpg", cv.IMREAD_GRAYSCALE)
+    f = np.fft.fft2(img)
+    fshift = np.fft.fftshift(f)
+    magnitude_spectrum0 = 20 * np.log(1 + np.abs(fshift))
+    plt.figure(figsize=(10, 5))
+    plt.subplot(151)
+    plt.imshow(img, cmap="gray")
+    plt.title("原图像")
+    plt.axis("off")
+    plt.subplot(152)
+    plt.imshow(magnitude_spectrum0, cmap="gray")
+    plt.title("原幅值谱")
+    plt.axis("off")
+    D0 = 20
+    n = 4
+    rows, cols = fshift.shape
+    crow, ccol = rows // 2, cols // 2
+    H = np.zeros((rows, cols))
+    for u in range(rows):
+        for v in range(cols):
+            D = sqrt((u - crow) ** 2 + (v - ccol) ** 2)
+            H[u, v] = 1 / (1 + (D / D0) ** (2 * n))
+    G = H * fshift
+    magnitude_spectrum1 = 20 * np.log(1 + np.abs(G))
+    f1 = np.fft.ifftshift(G)
+    img1 = np.abs(np.fft.ifft2(f1))
+    plt.subplot(153)
+    plt.imshow(H, cmap="gray")
+    plt.title("巴特沃斯传递函数")
+    plt.axis("off")
+    plt.subplot(154)
+    plt.imshow(magnitude_spectrum1, cmap="gray")
+    plt.title("巴特沃斯滤波后的幅值谱")
+    plt.axis("off")
+    plt.subplot(155)
+    plt.imshow(img1, cmap="gray")
+    plt.title("重构图像")
+    plt.axis("off")
+    plt.show()

3/4.py

@@ -0,0 +1,29 @@
+import cv2 as cv
+import numpy as np
+import matplotlib.pyplot as plt
+
+plt.rcParams["font.sans-serif"] = ["SimHei"]
+plt.rcParams["axes.unicode_minus"] = False
+
+img = cv.imread(r"img\peppers.bmp", 0)
+m, n = img.shape
+if m % 2 == 0:
+    m = m + 1
+if n % 2 == 0:
+    n = n + 1
+img = cv.resize(img, (m, n))
+f = np.fft.fft2(img)
+fshift = np.fft.fftshift(f)
+fmax = np.max(np.abs(fshift))
+u0, v0 = (m - 1) // 2, (n - 1) // 2
+u1 = u0 - 40
+v1 = v0 + 40
+fshift[v1, u1] = fmax / 5
+u2 = m - 1 - u1
+v2 = n - 1 - v1
+fshift[v2, u2] = fmax / 5
+f1 = np.fft.ifftshift(fshift)
+img1 = abs(np.fft.ifft2(f1))
+plt.imshow(img1, cmap="gray")
+plt.axis("off")
+plt.show()