需要用到scipy,安装不上可以换源
https://mirrors.aliyun.com/pypi/simple/
第137行选择原图,显示图像要等一会才能弹出来。
算法全部来自:这里
下面只加了点注释,更改了图片的显示方式,方便图片对比
import cv2
import numpy as np
import matplotlib.pyplot as plt
import scipy
import scipy.stats
import random
plt.rcParams['font.sans-serif'] = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False # 用来正常显示负号
def GaussieNoisy(image, sigma):
# 高斯噪声
img = image.astype(np.int16) # 此步是为了避免像素点小于0,大于255的情况
mu = 0
for i in range(img.shape[0]):
for j in range(img.shape[1]):
for k in range(img.shape[2]):
img[i, j, k] = img[i, j, k] + random.gauss(mu=mu, sigma=sigma)
img[img > 255] = 255
img[img < 0] = 0
img = img.astype(np.uint8)
return img
def spNoisy(image, s_vs_p=0.5, amount=0.004):
# 椒盐噪声
out = np.copy(image)
num_salt = np.ceil(amount * image.size * s_vs_p)
coords = [np.random.randint(0, i - 1, int(num_salt)) for i in image.shape]
out[tuple(coords)] = 1
num_pepper = np.ceil(amount * image.size * (1. - s_vs_p))
coords = [np.random.randint(0, i - 1, int(num_pepper)) for i in image.shape]
out[tuple(coords)] = 0
return out
def ArithmeticMeanAlogrithm(image):
# 算术均值滤波
new_image = np.zeros(image.shape)
image = cv2.copyMakeBorder(image, 1, 1, 1, 1, cv2.BORDER_DEFAULT)
for i in range(1, image.shape[0] - 1):
for j in range(1, image.shape[1] - 1):
new_image[i - 1, j - 1] = np.mean(image[i - 1:i + 2, j - 1:j + 2])
new_image = (new_image - np.min(image)) * (255 / np.max(image))
return new_image.astype(np.uint8)
def rgbArithmeticMean(image):
# 三个通道都进行算术均值滤波,最后合并
r, g, b = cv2.split(image)
r = ArithmeticMeanAlogrithm(r)
g = ArithmeticMeanAlogrithm(g)
b = ArithmeticMeanAlogrithm(b)
return cv2.merge([r, g, b])
def GeometricMeanOperator(roi):
roi = roi.astype(np.float64)
p = np.prod(roi)
return p ** (1 / (roi.shape[0] * roi.shape[1]))
def GeometricMeanAlogrithm(image):
# 几何均值滤波
new_image = np.zeros(image.shape)
image = cv2.copyMakeBorder(image, 1, 1, 1, 1, cv2.BORDER_DEFAULT)
for i in range(1, image.shape[0] - 1):
for j in range(1, image.shape[1] - 1):
new_image[i - 1, j - 1] = GeometricMeanOperator(image[i - 1:i + 2, j - 1:j + 2])
new_image = (new_image - np.min(image)) * (255 / np.max(image))
return new_image.astype(np.uint8)
def rgbGemotriccMean(image):
# 三个通道都进行几何均值滤波,最后合并
r, g, b = cv2.split(image)
r = GeometricMeanAlogrithm(r)
g = GeometricMeanAlogrithm(g)
b = GeometricMeanAlogrithm(b)
return cv2.merge([r, g, b])
def HarmonicMeanOperator(roi):
roi = roi.astype(np.float64)
if 0 in roi:
roi = 0
else:
roi = scipy.stats.hmean(roi.reshape(-1))
return roi
def HarmonicMeanAlogrithm(image):
# 谐波均值滤波
new_image = np.zeros(image.shape)
image = cv2.copyMakeBorder(image, 1, 1, 1, 1, cv2.BORDER_DEFAULT)
for i in range(1, image.shape[0] - 1):
for j in range(1, image.shape[1] - 1):
new_image[i - 1, j - 1] = HarmonicMeanOperator(image[i - 1:i + 2, j - 1:j + 2])
new_image = (new_image - np.min(image)) * (255 / np.max(image))
return new_image.astype(np.uint8)
def rgbHarmonicMean(image):
r, g, b = cv2.split(image)
r = HarmonicMeanAlogrithm(r)
g = HarmonicMeanAlogrithm(g)
b = HarmonicMeanAlogrithm(b)
return cv2.merge([r, g, b])
def Contra_harmonicMeanOperator(roi, q):
roi = roi.astype(np.float64)
return np.mean((roi) ** (q + 1)) / np.mean((roi) ** (q))
def Contra_harmonicMeanAlogrithm(image, q):
# 逆谐波均值滤波
new_image = np.zeros(image.shape)
image = cv2.copyMakeBorder(image, 1, 1, 1, 1, cv2.BORDER_DEFAULT)
for i in range(1, image.shape[0] - 1):
for j in range(1, image.shape[1] - 1):
new_image[i - 1, j - 1] = Contra_harmonicMeanOperator(image[i - 1:i + 2, j - 1:j + 2], q)
new_image = (new_image - np.min(image)) * (255 / np.max(image))
return new_image.astype(np.uint8)
def rgbContra_harmonicMean(image, q):
r, g, b = cv2.split(image)
r = Contra_harmonicMeanAlogrithm(r, q)
g = Contra_harmonicMeanAlogrithm(g, q)
b = Contra_harmonicMeanAlogrithm(b, q)
return cv2.merge([r, g, b])
if __name__ == '__main__':
img = cv2.imread("img/lena.png")
img = cv2.resize(cv2.cvtColor(img, cv2.COLOR_BGR2RGB), (200, 200))
img_gaussie_noisy = GaussieNoisy(img, 18)
img_sp_noisy = spNoisy(img)
plt.subplot(161), plt.imshow(img), plt.title("原图"), plt.xticks([]), plt.yticks([])
plt.subplot(162), plt.imshow(img_gaussie_noisy), plt.title("加高斯噪声"), plt.xticks(
[]), plt.yticks([])
plt.subplot(163), plt.imshow(rgbArithmeticMean(img_gaussie_noisy)), plt.title("算数均值"), plt.xticks(
[]), plt.yticks([])
plt.subplot(164), plt.imshow(rgbGemotriccMean(img_gaussie_noisy)), plt.title("几何均值"), plt.xticks(
[]), plt.yticks([])
plt.subplot(165), plt.imshow(rgbHarmonicMean(img_gaussie_noisy)), plt.title("谐波"), plt.xticks(
[]), plt.yticks([])
plt.subplot(166), plt.imshow(rgbContra_harmonicMean(img_gaussie_noisy, 2)), plt.title("逆谐波"), plt.xticks(
[]), plt.yticks([])
plt.show()
plt.subplot(161), plt.imshow(img), plt.title("原图"), plt.xticks([]), plt.yticks([])
plt.subplot(162), plt.imshow(img_sp_noisy), plt.title("加椒盐噪声"), plt.xticks(
[]), plt.yticks([])
plt.subplot(163), plt.imshow(rgbArithmeticMean(img_sp_noisy)), plt.title("算数均值"), plt.xticks(
[]), plt.yticks([])
plt.subplot(164), plt.imshow(rgbGemotriccMean(img_sp_noisy)), plt.title("几何均值"), plt.xticks(
[]), plt.yticks([])
plt.subplot(165), plt.imshow(rgbHarmonicMean(img_sp_noisy)), plt.title("谐波"), plt.xticks(
[]), plt.yticks([])
plt.subplot(166), plt.imshow(rgbContra_harmonicMean(img_sp_noisy, 2)), plt.title("逆谐波"), plt.xticks(
[]), plt.yticks([])
plt.show()