导向滤波的python代码

导向滤波是一种基于图像的非线性滤波方法，可以在保持边缘信息的同时对图像进行平滑处理。下面是一个简单的 Python 实现：

import numpy as np
import cv2

def guided_filter(I, p, r, eps):
    """
    :param I: guidance image (should be a gray-scale/single channel image)
    :param p: filtering input image (should be a gray-scale/single channel image)
    :param r: local window radius
    :param eps: regularization parameter
    :return: filtering output q
    """
    # Step 1: mean of I, p and I*p
    mean_I = cv2.boxFilter(I, cv2.CV_64F, (r,r))
    mean_p = cv2.boxFilter(p, cv2.CV_64F, (r,r))
    mean_Ip = cv2.boxFilter(I*p, cv2.CV_64F, (r,r))

    # Step 2: covariance of I and p
    cov_Ip = mean_Ip - mean_I * mean_p

    # Step 3: mean and variance of I
    mean_II = cv2.boxFilter(I*I, cv2.CV_64F, (r,r))
    var_I = mean_II - mean_I * mean_I

    # Step 4: alpha and beta
    a = cov_Ip / (var_I + eps)
    b = mean_p - a * mean_I

    # Step 5: mean of a and b
    mean_a = cv2.boxFilter(a, cv2.CV_64F, (r,r))
    mean_b = cv2.boxFilter(b, cv2.CV_64F, (r,r))

    # Step 6: filtering output q
    q = mean_a * I + mean_b
    return q.astype(np.uint8)

其中，`I` 是导向图像，`p` 是需要进行滤波的图像，`r` 是局部窗口半径，`eps` 是正则化参数。函数返回滤波后的图像 `q`。