One hundred questions of image processing (11-20)

Question 11 ： Mean filter

Use 3*3 Let's use the average filter of ！

The output of the smooth linear spatial filter is a simple average of the pixels contained in the neighborhood of the filter template , Mean filter . The mean filter is also a low-pass filter , Mean filter is easy to understand , That is, assign the average value in the neighborhood to the central element .
Mean filter is used to reduce noise , The main application of mean filter is to remove irrelevant details in the image , Uncorrelation refers to the small pixel area compared with the filter template . Blur the picture to get a rough description of the object of interest , So the grayscale of those smaller objects will be mixed with the background , Larger objects become like spots and are easy to detect . The size of the template is determined by the size of the objects that will blend into the background .

import cv2
import numpy as np

# mean filter
def mean_filter(img, K_size=3):
    H, W, C = img.shape

    # zero padding
    pad = K_size // 2
    out = np.zeros((H + pad * 2, W + pad * 2, C), dtype=np.float)
    out[pad: pad + H, pad: pad + W] = img.copy().astype(np.float)
    tmp = out.copy()

    # filtering
    for y in range(H):
        for x in range(W):
            for c in range(C):
                out[pad + y, pad + x, c] = np.mean(tmp[y: y + K_size, x: x + K_size, c])

    out = out[pad: pad + H, pad: pad + W].astype(np.uint8)

    return out

# Read image
img = cv2.imread("jiaoyan.jpg")

# Mean Filter
out = mean_filter(img, K_size=3)

# Save result
cv2.imwrite("out.jpg", out)
cv2.imshow("result", out)
cv2.waitKey(0)
cv2.destroyAllWindows()

Original image

After processing the image

Question 12 ：Motion Filter

Motion Filter Take the average value of pixels in the diagonal direction , Define as follows ：

import cv2
import numpy as np

# motion filter
def motion_filter(img, K_size=3):
    H, W, C = img.shape

    # Kernel
    K = np.diag( [1] * K_size ).astype(np.float)
    K /= K_size

    # zero padding
    pad = K_size // 2
    out = np.zeros((H + pad * 2, W + pad * 2, C), dtype=np.float)
    out[pad: pad + H, pad: pad + W] = img.copy().astype(np.float)
    tmp = out.copy()

    # filtering
    for y in range(H):
        for x in range(W):
            for c in range(C):
                out[pad + y, pad + x, c] = np.sum(K * tmp[y: y + K_size, x: x + K_size, c])

    out = out[pad: pad + H, pad: pad + W].astype(np.uint8)

    return out


# Read image
img = cv2.imread("jiaoyan.jpg")

# motion filtering
out = motion_filter(img, K_size=3)

# Save result
cv2.imwrite("out.jpg", out)
cv2.imshow("result", out)
cv2.waitKey(0)
cv2.destroyAllWindows()