Noise Removal

import cv2
import numpy as np

image = cv2.imread("photo.jpg")
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

# --- ADD ARTIFICIAL NOISE FOR DEMONSTRATION ---

# Salt-and-pepper noise (random black/white pixels)
def add_salt_pepper(img, amount=0.02):
    noisy = img.copy()
    n_pixels = int(amount * img.size)
    # Salt (white)
    coords = [np.random.randint(0, i-1, n_pixels) for i in img.shape]
    noisy[coords[0], coords[1]] = 255
    # Pepper (black)
    coords = [np.random.randint(0, i-1, n_pixels) for i in img.shape]
    noisy[coords[0], coords[1]] = 0
    return noisy

# Gaussian noise (random intensity variation)
def add_gaussian_noise(img, std=25):
    noise = np.random.normal(0, std, img.shape).astype(np.int16)
    noisy = np.clip(img.astype(np.int16) + noise, 0, 255).astype(np.uint8)
    return noisy

sp_noisy = add_salt_pepper(gray)
gauss_noisy = add_gaussian_noise(gray)

# --- REMOVE NOISE ---

# Salt-and-pepper → use Median blur
sp_denoised = cv2.medianBlur(sp_noisy, 5)

# Gaussian noise → use Gaussian blur or Non-local Means
gauss_denoised_gb = cv2.GaussianBlur(gauss_noisy, (5, 5), 0)
gauss_denoised_nlm = cv2.fastNlMeansDenoising(
    gauss_noisy, h=10, templateWindowSize=7, searchWindowSize=21
)

cv2.imwrite("sp_noisy.jpg", sp_noisy)
cv2.imwrite("sp_denoised.jpg", sp_denoised)
cv2.imwrite("gauss_denoised.jpg", gauss_denoised_nlm)
print("Noise removal complete")