环境:Python3.8 和 OpenCV
内容:图像的高斯滤波和双边滤波
双边滤波:
缺点:不适合处理椒盐噪声
优点:保留边缘等高频信息,平滑灰度相近的区域,综合考虑距离和颜色差异两种因素
import cv2 as cv
import numpy as np
import matplotlib.pyplot as plt
# 封装图片显示函数
def image_show(image):
if image.ndim == 2:
plt.imshow(image, cmap='gray')
else:
image = cv.cvtColor(image, cv.COLOR_BGR2RGB)
plt.imshow(image)
plt.show()
# 封装添加椒盐噪声函数(应输入灰度图)
def sp_noise(image, proportion):
# 获取添加噪声的比例大小
[height, width] = image.shape
num = int(height * width * proportion)
# 按照比例对图像添加椒盐噪声
for i in range(num):
w = np.random.randint(0, width - 1)
h = np.random.randint(0, height - 1)
if np.random.randint(0, 2) == 0:
image[h, w] = 0
else:
image[h, w] = 255
return image
if __name__ == '__main__':
# 读取灰度图像
img_desk = cv.imread('desk.png', 0)
image_show(img_desk)
# 添加椒盐噪声
prop = 0.05 # 添加椒盐噪声的比例
img_noise = sp_noise(img_desk, prop)
image_show(img_noise)
# 高斯滤波
sigma = 5 # 高斯函数的sigma参数
kernel = [5, 5] # 卷积核大小
img_gaussian = cv.GaussianBlur(img_noise, kernel, sigmaX=sigma)
image_show(img_gaussian)
# 双边滤波
sigma_color = 50 # 颜色差异卷积核参数
sigma_space = 3 # 距离差异卷积核参数
img_bilateral = cv.bilateralFilter(img_noise, -1, sigmaColor=sigma_color, sigmaSpace=sigma_space)
image_show(img_bilateral)