opencv-python实现 图像增强

利用频率域滤波方法去掉图像“house”中的竖条纹

import cv2
import numpy as np
from matplotlib import pyplot as plt

# 将gif图片转换成可识别图像
gif = cv2.VideoCapture('house.gif')
ret, frame = gif.read()
if ret:
    print('change success!')
    img = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
else:
    print("change fail!")

# 傅里叶变换
dft = cv2.dft(np.float32(img), flags=cv2.DFT_COMPLEX_OUTPUT)  # 将空间域转换为频率域
dft_shift = np.fft.fftshift(dft)  # 将低频部分移动到图像中心
rows, cols = img.shape

# create a mask first, center square is 1, remaining all zeros
mask = np.zeros((rows, cols, 2), np.uint8)
for i in range(rows):
    for j in range(cols):
        mask[i][j] = 1

m = 18 # 越小越清晰，越大越糊
n = 14 # 越小越糊，越大越清晰
mask[int(rows / 2 - m):int(rows / 2 + m), int(cols / 2 - m):int(cols / 2 + m)] = 0
mask[int(rows / 2 - n):int(rows / 2 + n), int(cols / 2 - n):int(cols / 2 + n)] = 1

# apply mask and inverse DFT
fshift = dft_shift * mask


f_ishift = np.fft.ifftshift(fshift) # fftshit()函数的逆函数，它将频谱图像的中心低频部分移动至左上角
img_back = cv2.idft(f_ishift) # 将频率域转化回空间域，输出是一个复数，cv2.idft()返回的是一个双通道图像
img_back2 = cv2.magnitude(img_back[:, :, 0], img_back[:, :, 1]) # idft[:,:,0]求得实部，用idft[:,:,1]求得虚部。

plt.subplot(131), plt.imshow(img, cmap='gray')
plt.title('Input Image'), plt.xticks([]), plt.yticks([])
plt.subplot(132), plt.imshow(img_back2, cmap='gray')
plt.title('Magnitude Spectrum'), plt.xticks([]), plt.yticks([])
plt.subplot(133), plt.imshow(20*np.log(img_back2), cmap='gray') # 归一化图像
plt.title('normalize'), plt.xticks([]), plt.yticks([])
plt.show()