Computer vision: Median filter with replicated borders

The median filter wipes out salt-and-pepper noise and keeps edges sharp.

What to build

A program that applies a k × k median filter to every pixel.

Requirements

  • Each pixel becomes the middle value of the sorted k × k window centred on it.
  • Outside the image, use replicate padding: copy the nearest edge pixel.
  • Print H rows of W values.

Input

Line 1: H, W and k (odd). Then H rows of W values.

Output

H rows.

Constraints

1 ≤ H, W ≤ 40; k is 1, 3 or 5.

Example

Input:

3 3 3
10 10 10
10 255 10
10 10 10

Output:

10 10 10
10 10 10
10 10 10

Why it works

Extreme values fall to the ends of the sorted window, so the middle value comes from the majority of the neighbourhood; clamping the indices gives border pixels a full window too.

Examples

Example 1 · Input
3 3 3
10 10 10
10 255 10
10 10 10
Expected output
10 10 10
10 10 10
10 10 10
Example 2 · Input
1 5 3
5 0 5 9 5
Expected output
5 5 5 5 5

Prompts inside input() are ignored when checking, so input("n: ") is fine.

Hints

Hint 1

For pixel (r, c), read rows r − m to r + m and columns c − m to c + m with m = k // 2, clamping each index into the image.

Hint 2

Sort the k × k values; the median is at index (k × k) // 2.

Solution

Show one correct solution (try it yourself first)
h, w, k = map(int, input().split())
img = [list(map(int, input().split())) for _ in range(h)]
m = k // 2
for r in range(h):
    out = []
    for c in range(w):
        win = sorted(img[min(max(r + i, 0), h - 1)][min(max(c + j, 0), w - 1)] for i in range(-m, m + 1) for j in range(-m, m + 1))
        out.append(win[len(win) // 2])
    print(" ".join(map(str, out)))

Lesson: Blurring and noise removal →